Legged_paper_daily_arxiv
Robotics 51
☆ 3P-LLM: Probabilistic Path Planning using Large Language Model for Autonomous Robot Navigation
Much worldly semantic knowledge can be encoded in large language models (LLMs). Such information could be of great use to robots that want to carry out high-level, temporally extended commands stated in natural language. However, the lack of real-world experience that language models have is a key limitation that makes it challenging to use them for decision-making inside a particular embodiment. This research assesses the feasibility of using LLM (GPT-3.5-turbo chatbot by OpenAI) for robotic path planning. The shortcomings of conventional approaches to managing complex environments and developing trustworthy plans for shifting environmental conditions serve as the driving force behind the research. Due to the sophisticated natural language processing abilities of LLM, the capacity to provide effective and adaptive path-planning algorithms in real-time, great accuracy, and few-shot learning capabilities, GPT-3.5-turbo is well suited for path planning in robotics. In numerous simulated scenarios, the research compares the performance of GPT-3.5-turbo with that of state-of-the-art path planners like Rapidly Exploring Random Tree (RRT) and A*. We observed that GPT-3.5-turbo is able to provide real-time path planning feedback to the robot and outperforms its counterparts. This paper establishes the foundation for LLM-powered path planning for robotic systems.
comment: Exploratory Study
☆ CaT: Constraints as Terminations for Legged Locomotion Reinforcement Learning
Deep Reinforcement Learning (RL) has demonstrated impressive results in solving complex robotic tasks such as quadruped locomotion. Yet, current solvers fail to produce efficient policies respecting hard constraints. In this work, we advocate for integrating constraints into robot learning and present Constraints as Terminations (CaT), a novel constrained RL algorithm. Departing from classical constrained RL formulations, we reformulate constraints through stochastic terminations during policy learning: any violation of a constraint triggers a probability of terminating potential future rewards the RL agent could attain. We propose an algorithmic approach to this formulation, by minimally modifying widely used off-the-shelf RL algorithms in robot learning (such as Proximal Policy Optimization). Our approach leads to excellent constraint adherence without introducing undue complexity and computational overhead, thus mitigating barriers to broader adoption. Through empirical evaluation on the real quadruped robot Solo crossing challenging obstacles, we demonstrate that CaT provides a compelling solution for incorporating constraints into RL frameworks. Videos and code are available at https://constraints-as-terminations.github.io.
comment: Project webpage: https://constraints-as-terminations.github.io
☆ Temporal Logic Formalisation of ISO 34502 Critical Scenarios: Modular Construction with the RSS Safety Distance
As the development of autonomous vehicles progresses, efficient safety assurance methods become increasingly necessary. Safety assurance methods such as monitoring and scenario-based testing call for formalisation of driving scenarios. In this paper, we develop a temporal-logic formalisation of an important class of critical scenarios in the ISO standard 34502. We use signal temporal logic (STL) as a logical formalism. Our formalisation has two main features: 1) modular composition of logical formulas for systematic and comprehensive formalisation (following the compositional methodology of ISO 34502); 2) use of the RSS distance for defining danger. We find our formalisation comes with few parameters to tune thanks to the RSS distance. We experimentally evaluated our formalisation; using its results, we discuss the validity of our formalisation and its stability with respect to the choice of some parameter values.
comment: 12 pages, 4 figures, 5 tables. Accepted to SAC 2024
☆ ModaLink: Unifying Modalities for Efficient Image-to-PointCloud Place Recognition
Place recognition is an important task for robots and autonomous cars to localize themselves and close loops in pre-built maps. While single-modal sensor-based methods have shown satisfactory performance, cross-modal place recognition that retrieving images from a point-cloud database remains a challenging problem. Current cross-modal methods transform images into 3D points using depth estimation for modality conversion, which are usually computationally intensive and need expensive labeled data for depth supervision. In this work, we introduce a fast and lightweight framework to encode images and point clouds into place-distinctive descriptors. We propose an effective Field of View (FoV) transformation module to convert point clouds into an analogous modality as images. This module eliminates the necessity for depth estimation and helps subsequent modules achieve real-time performance. We further design a non-negative factorization-based encoder to extract mutually consistent semantic features between point clouds and images. This encoder yields more distinctive global descriptors for retrieval. Experimental results on the KITTI dataset show that our proposed methods achieve state-of-the-art performance while running in real time. Additional evaluation on the HAOMO dataset covering a 17 km trajectory further shows the practical generalization capabilities. We have released the implementation of our methods as open source at: https://github.com/haomo-ai/ModaLink.git.
comment: 8 pages, 11 figures, conference
☆ MLDT: Multi-Level Decomposition for Complex Long-Horizon Robotic Task Planning with Open-Source Large Language Model
In the realm of data-driven AI technology, the application of open-source large language models (LLMs) in robotic task planning represents a significant milestone. Recent robotic task planning methods based on open-source LLMs typically leverage vast task planning datasets to enhance models' planning abilities. While these methods show promise, they struggle with complex long-horizon tasks, which require comprehending more context and generating longer action sequences. This paper addresses this limitation by proposing MLDT, theMulti-Level Decomposition Task planning method. This method innovatively decomposes tasks at the goal-level, task-level, and action-level to mitigate the challenge of complex long-horizon tasks. In order to enhance open-source LLMs' planning abilities, we introduce a goal-sensitive corpus generation method to create high-quality training data and conduct instruction tuning on the generated corpus. Since the complexity of the existing datasets is not high enough, we construct a more challenging dataset, LongTasks, to specifically evaluate planning ability on complex long-horizon tasks. We evaluate our method using various LLMs on four datasets in VirtualHome. Our results demonstrate a significant performance enhancement in robotic task planning, showcasing MLDT's effectiveness in overcoming the limitations of existing methods based on open-source LLMs as well as its practicality in complex, real-world scenarios.
☆ PhysicsAssistant: An LLM-Powered Interactive Learning Robot for Physics Lab Investigations
Robot systems in education can leverage Large language models' (LLMs) natural language understanding capabilities to provide assistance and facilitate learning. This paper proposes a multimodal interactive robot (PhysicsAssistant) built on YOLOv8 object detection, cameras, speech recognition, and chatbot using LLM to provide assistance to students' physics labs. We conduct a user study on ten 8th-grade students to empirically evaluate the performance of PhysicsAssistant with a human expert. The Expert rates the assistants' responses to student queries on a 0-4 scale based on Bloom's taxonomy to provide educational support. We have compared the performance of PhysicsAssistant (YOLOv8+GPT-3.5-turbo) with GPT-4 and found that the human expert rating of both systems for factual understanding is the same. However, the rating of GPT-4 for conceptual and procedural knowledge (3 and 3.2 vs 2.2 and 2.6, respectively) is significantly higher than PhysicsAssistant (p < 0.05). However, the response time of GPT-4 is significantly higher than PhysicsAssistant (3.54 vs 1.64 sec, p < 0.05). Hence, despite the relatively lower response quality of PhysicsAssistant than GPT-4, it has shown potential for being used as a real-time lab assistant to provide timely responses and can offload teachers' labor to assist with repetitive tasks. To the best of our knowledge, this is the first attempt to build such an interactive multimodal robotic assistant for K-12 science (physics) education.
comment: Submitted to IEEE RO-MAN
☆ An Efficient Risk-aware Branch MPC for Automated Driving that is Robust to Uncertain Vehicle Behaviors
One of the critical challenges in automated driving is ensuring safety of automated vehicles despite the unknown behavior of the other vehicles. Although motion prediction modules are able to generate a probability distribution associated with various behavior modes, their probabilistic estimates are often inaccurate, thus leading to a possibly unsafe trajectory. To overcome this challenge, we propose a risk-aware motion planning framework that appropriately accounts for the ambiguity in the estimated probability distribution. We formulate the risk-aware motion planning problem as a min-max optimization problem and develop an efficient iterative method by incorporating a regularization term in the probability update step. Via extensive numerical studies, we validate the convergence of our method and demonstrate its advantages compared to the state-of-the-art approaches.
☆ Teaching Introductory HRI: UChicago Course "Human-Robot Interaction: Research and Practice"
In 2020, I designed the course CMSC 20630/30630 Human-Robot Interaction: Research and Practice as a hands-on introduction to human-robot interaction (HRI) research for both undergraduate and graduate students at the University of Chicago. Since 2020, I have taught and refined this course each academic year. Human-Robot Interaction: Research and Practice focuses on the core concepts and cutting-edge research in the field of human-robot interaction (HRI), covering topics that include: nonverbal robot behavior, verbal robot behavior, social dynamics, norms & ethics, collaboration & learning, group interactions, applications, and future challenges of HRI. Course meetings involve students in the class leading discussions about cutting-edge peer-reviewed research HRI publications. Students also participate in a quarter-long collaborative research project, where they pursue an HRI research question that often involves conducing their own human-subjects research study where they recruit human subjects to interact with a robot. In this paper, I detail the structure of the course and its learning goals as well as my reflections and student feedback on the course.
comment: 4 pages, 2 tables, Presented at the Designing an Intro to HRI Course Workshop at HRI 2024 (arXiv:2403.05588)
☆ Sampling-Based Motion Planning with Online Racing Line Generation for Autonomous Driving on Three-Dimensional Race Tracks
Existing approaches to trajectory planning for autonomous racing employ sampling-based methods, generating numerous jerk-optimal trajectories and selecting the most favorable feasible trajectory based on a cost function penalizing deviations from an offline-calculated racing line. While successful on oval tracks, these methods face limitations on complex circuits due to the simplistic geometry of jerk-optimal edges failing to capture the complexity of the racing line. Additionally, they only consider two-dimensional tracks, potentially neglecting or surpassing the actual dynamic potential. In this paper, we present a sampling-based local trajectory planning approach for autonomous racing that can maintain the lap time of the racing line even on complex race tracks and consider the race track's three-dimensional effects. In simulative experiments, we demonstrate that our approach achieves lower lap times and improved utilization of dynamic limits compared to existing approaches. We also investigate the impact of online racing line generation, in which the time-optimal solution is planned from the current vehicle state for a limited spatial horizon, in contrast to a closed racing line calculated offline. We show that combining the sampling-based planner with the online racing line generation can significantly reduce lap times in multi-vehicle scenarios.
comment: 8 pages, submitted to be published at the 35th IEEE Intelligent Vehicles Symposium, June 2 - 5, 2024, Jeju Shinhwa World, Jeju Island, Korea
☆ Will You Participate? Exploring the Potential of Robotics Competitions on Human-centric Topics
This paper presents findings from an exploratory needfinding study investigating the research current status and potential participation of the competitions on the robotics community towards four human-centric topics: safety, privacy, explainability, and federated learning. We conducted a survey with 34 participants across three distinguished European robotics consortia, nearly 60% of whom possessed over five years of research experience in robotics. Our qualitative and quantitative analysis revealed that current mainstream robotic researchers prioritize safety and explainability, expressing a greater willingness to invest in further research in these areas. Conversely, our results indicate that privacy and federated learning garner less attention and are perceived to have lower potential. Additionally, the study suggests a lack of enthusiasm within the robotics community for participating in competitions related to these topics. Based on these findings, we recommend targeting other communities, such as the machine learning community, for future competitions related to these four human-centric topics.
☆ RAP: Retrieval-Augmented Planner for Adaptive Procedure Planning in Instructional Videos
Procedure Planning in instructional videos entails generating a sequence of action steps based on visual observations of the initial and target states. Despite the rapid progress in this task, there remain several critical challenges to be solved: (1) Adaptive procedures: Prior works hold an unrealistic assumption that the number of action steps is known and fixed, leading to non-generalizable models in real-world scenarios where the sequence length varies. (2) Temporal relation: Understanding the step temporal relation knowledge is essential in producing reasonable and executable plans. (3) Annotation cost: Annotating instructional videos with step-level labels (i.e., timestamp) or sequence-level labels (i.e., action category) is demanding and labor-intensive, limiting its generalizability to large-scale datasets.In this work, we propose a new and practical setting, called adaptive procedure planning in instructional videos, where the procedure length is not fixed or pre-determined. To address these challenges we introduce Retrieval-Augmented Planner (RAP) model. Specifically, for adaptive procedures, RAP adaptively determines the conclusion of actions using an auto-regressive model architecture. For temporal relation, RAP establishes an external memory module to explicitly retrieve the most relevant state-action pairs from the training videos and revises the generated procedures. To tackle high annotation cost, RAP utilizes a weakly-supervised learning manner to expand the training dataset to other task-relevant, unannotated videos by generating pseudo labels for action steps. Experiments on CrossTask and COIN benchmarks show the superiority of RAP over traditional fixed-length models, establishing it as a strong baseline solution for adaptive procedure planning.
comment: 23 pages, 6 figures, 12 tables
☆ Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes
Fast and robust object grasping in clutter is a crucial component of robotics. Most current works resort to the whole observed point cloud for 6-Dof grasp generation, ignoring the guidance information excavated from global semantics, thus limiting high-quality grasp generation and real-time performance. In this work, we show that the widely used heatmaps are underestimated in the efficiency of 6-Dof grasp generation. Therefore, we propose an effective local grasp generator combined with grasp heatmaps as guidance, which infers in a global-to-local semantic-to-point way. Specifically, Gaussian encoding and the grid-based strategy are applied to predict grasp heatmaps as guidance to aggregate local points into graspable regions and provide global semantic information. Further, a novel non-uniform anchor sampling mechanism is designed to improve grasp accuracy and diversity. Benefiting from the high-efficiency encoding in the image space and focusing on points in local graspable regions, our framework can perform high-quality grasp detection in real-time and achieve state-of-the-art results. In addition, real robot experiments demonstrate the effectiveness of our method with a success rate of 94% and a clutter completion rate of 100%. Our code is available at https://github.com/THU-VCLab/HGGD.
comment: Extensive results on GraspNet-1B dataset
☆ Bridging the Gap: Regularized Reinforcement Learning for Improved Classical Motion Planning with Safety Modules
Classical navigation planners can provide safe navigation, albeit often suboptimally and with hindered human norm compliance. ML-based, contemporary autonomous navigation algorithms can imitate more natural and humancompliant navigation, but usually require large and realistic datasets and do not always provide safety guarantees. We present an approach that leverages a classical algorithm to guide reinforcement learning. This greatly improves the results and convergence rate of the underlying RL algorithm and requires no human-expert demonstrations to jump-start the process. Additionally, we incorporate a practical fallback system that can switch back to a classical planner to ensure safety. The outcome is a sample efficient ML approach for mobile navigation that builds on classical algorithms, improves them to ensure human compliance, and guarantees safety.
comment: 8 pages
☆ CoBOS: Constraint-Based Online Scheduler for Human-Robot Collaboration
Assembly processes involving humans and robots are challenging scenarios because the individual activities and access to shared workspace have to be coordinated. Fixed robot programs leave no room to diverge from a fixed protocol. Working on such a process can be stressful for the user and lead to ineffective behavior or failure. We propose a novel approach of online constraint-based scheduling in a reactive execution control framework facilitating behavior trees called CoBOS. This allows the robot to adapt to uncertain events such as delayed activity completions and activity selection (by the human). The user will experience less stress as the robotic coworkers adapt their behavior to best complement the human-selected activities to complete the common task. In addition to the improved working conditions, our algorithm leads to increased efficiency, even in highly uncertain scenarios. We evaluate our algorithm using a probabilistic simulation study with 56000 experiments. We outperform all baselines by a margin of 4-10%. Initial real robot experiments using a Franka Emika Panda robot and human tracking based on HTC Vive VR gloves look promising.
comment: 7 pages, 8 figures
☆ Inverse kinematics learning of a continuum manipulator using limited real time data
Data driven control of a continuum manipulator requires a lot of data for training but generating sufficient amount of real time data is not cost efficient. Random actuation of the manipulator can also be unsafe sometimes. Meta learning has been used successfully to adapt to a new environment. Hence, this paper tries to solve the above mentioned problem using meta learning. We consider two cases for that. First, this paper proposes a method to use simulation data for training the model using MAML(Model-Agnostic Meta-Learning). Then, it adapts to the real world using gradient steps. Secondly,if the simulation model is not available or difficult to formulate, then we propose a CGAN(Conditional Generative adversial network)-MAML based method for it. The model is trained using a small amount of real time data and augmented data for different loading conditions. Then, adaptation is done in the real environment. It has been found out from the experiments that the relative positioning error for both the cases are below 3%. The proposed models are experimentally verified on a real continuum manipulator.
☆ SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model CVPR 2024
There are five types of trajectory prediction tasks: deterministic, stochastic, domain adaptation, momentary observation, and few-shot. These associated tasks are defined by various factors, such as the length of input paths, data split and pre-processing methods. Interestingly, even though they commonly take sequential coordinates of observations as input and infer future paths in the same coordinates as output, designing specialized architectures for each task is still necessary. For the other task, generality issues can lead to sub-optimal performances. In this paper, we propose SingularTrajectory, a diffusion-based universal trajectory prediction framework to reduce the performance gap across the five tasks. The core of SingularTrajectory is to unify a variety of human dynamics representations on the associated tasks. To do this, we first build a Singular space to project all types of motion patterns from each task into one embedding space. We next propose an adaptive anchor working in the Singular space. Unlike traditional fixed anchor methods that sometimes yield unacceptable paths, our adaptive anchor enables correct anchors, which are put into a wrong location, based on a traversability map. Finally, we adopt a diffusion-based predictor to further enhance the prototype paths using a cascaded denoising process. Our unified framework ensures the generality across various benchmark settings such as input modality, and trajectory lengths. Extensive experiments on five public benchmarks demonstrate that SingularTrajectory substantially outperforms existing models, highlighting its effectiveness in estimating general dynamics of human movements. Code is publicly available at https://github.com/inhwanbae/SingularTrajectory .
comment: Accepted at CVPR 2024
☆ Can Language Beat Numerical Regression? Language-Based Multimodal Trajectory Prediction CVPR 2024
Language models have demonstrated impressive ability in context understanding and generative performance. Inspired by the recent success of language foundation models, in this paper, we propose LMTraj (Language-based Multimodal Trajectory predictor), which recasts the trajectory prediction task into a sort of question-answering problem. Departing from traditional numerical regression models, which treat the trajectory coordinate sequence as continuous signals, we consider them as discrete signals like text prompts. Specially, we first transform an input space for the trajectory coordinate into the natural language space. Here, the entire time-series trajectories of pedestrians are converted into a text prompt, and scene images are described as text information through image captioning. The transformed numerical and image data are then wrapped into the question-answering template for use in a language model. Next, to guide the language model in understanding and reasoning high-level knowledge, such as scene context and social relationships between pedestrians, we introduce an auxiliary multi-task question and answering. We then train a numerical tokenizer with the prompt data. We encourage the tokenizer to separate the integer and decimal parts well, and leverage it to capture correlations between the consecutive numbers in the language model. Lastly, we train the language model using the numerical tokenizer and all of the question-answer prompts. Here, we propose a beam-search-based most-likely prediction and a temperature-based multimodal prediction to implement both deterministic and stochastic inferences. Applying our LMTraj, we show that the language-based model can be a powerful pedestrian trajectory predictor, and outperforms existing numerical-based predictor methods. Code is publicly available at https://github.com/inhwanbae/LMTrajectory .
comment: Accepted at CVPR 2024
☆ HyRRT-Connect: A Bidirectional Rapidly-Exploring Random Trees Motion Planning Algorithm for Hybrid Systems
This paper proposes a bidirectional rapidly-exploring random trees (RRT) algorithm to solve the motion planning problem for hybrid systems. The proposed algorithm, called HyRRT-Connect, propagates in both forward and backward directions in hybrid time until an overlap between the forward and backward propagation results is detected. Then, HyRRT-Connect constructs a motion plan through the reversal and concatenation of functions defined on hybrid time domains, ensuring the motion plan thoroughly satisfies the given hybrid dynamics. To address the potential discontinuity along the flow caused by tolerating some distance between the forward and backward partial motion plans, we reconstruct the backward partial motion plan by a forward-in-hybrid-time simulation from the final state of the forward partial motion plan. By applying the reversed input of the backward partial motion plan, the reconstruction process effectively eliminates the discontinuity and ensures that as the tolerance distance decreases to zero, the distance between the endpoint of the reconstructed motion plan and the final state set approaches zero. The proposed algorithm is applied to an actuated bouncing ball example and a walking robot example so as to highlight its generality and computational improvement.
comment: Accepted by the 8th IFAC International Conference on Analysis and Design of Hybrid Systems (ADHS 2024)
☆ Extensible Hook System for Rendesvouz and Docking of a Cubesat Swarm
The use of cubesat swarms is being proposed for different missions where cooperation between satellites is required. Commonly, the cube swarm requires formation flight and even rendezvous and docking, which are very challenging tasks since they required more energy and the use of advanced guidance, navigation and control techniques. In this paper, we propose the use of an extensible hook system to mitigate these drawbacks,i.e. it allows to save fuel and reduce the system complexity by including techniques that have been previously demonstrated on Earth. This system is based on a scissor boom structure, which could reach up to five meters for a 4U dimension, including three degrees of freedom to place the end effector at any pose within the system workspace. We simulated the dynamic behaviour of a cubesat with the proposed system, demonstrating the required power for a 16U cubesat equipped with one extensible hook system is considered acceptable according to the current state of the art actuators.
☆ Imaging radar and LiDAR image translation for 3-DOF extrinsic calibration
The integration of sensor data is crucial in the field of robotics to take full advantage of the various sensors employed. One critical aspect of this integration is determining the extrinsic calibration parameters, such as the relative transformation, between each sensor. The use of data fusion between complementary sensors, such as radar and LiDAR, can provide significant benefits, particularly in harsh environments where accurate depth data is required. However, noise included in radar sensor data can make the estimation of extrinsic calibration challenging. To address this issue, we present a novel framework for the extrinsic calibration of radar and LiDAR sensors, utilizing CycleGAN as amethod of image-to-image translation. Our proposed method employs translating radar bird-eye-view images into LiDAR-style images to estimate the 3-DOF extrinsic parameters. The use of image registration techniques, as well as deskewing based on sensor odometry and B-spline interpolation, is employed to address the rolling shutter effect commonly present in spinning sensors. Our method demonstrates a notable improvement in extrinsic calibration compared to filter-based methods using the MulRan dataset.
☆ RoboKeyGen: Robot Pose and Joint Angles Estimation via Diffusion-based 3D Keypoint Generation ICRA 2024
Estimating robot pose and joint angles is significant in advanced robotics, enabling applications like robot collaboration and online hand-eye calibration.However, the introduction of unknown joint angles makes prediction more complex than simple robot pose estimation, due to its higher dimensionality.Previous methods either regress 3D keypoints directly or utilise a render&compare strategy. These approaches often falter in terms of performance or efficiency and grapple with the cross-camera gap problem.This paper presents a novel framework that bifurcates the high-dimensional prediction task into two manageable subtasks: 2D keypoints detection and lifting 2D keypoints to 3D. This separation promises enhanced performance without sacrificing the efficiency innate to keypoint-based techniques.A vital component of our method is the lifting of 2D keypoints to 3D keypoints. Common deterministic regression methods may falter when faced with uncertainties from 2D detection errors or self-occlusions.Leveraging the robust modeling potential of diffusion models, we reframe this issue as a conditional 3D keypoints generation task. To bolster cross-camera adaptability, we introduce theNormalised Camera Coordinate Space (NCCS), ensuring alignment of estimated 2D keypoints across varying camera intrinsics.Experimental results demonstrate that the proposed method outperforms the state-of-the-art render\&compare method and achieves higher inference speed.Furthermore, the tests accentuate our method's robust cross-camera generalisation capabilities.We intend to release both the dataset and code in https://nimolty.github.io/Robokeygen/
comment: Accepted by ICRA 2024
☆ Manipulating Neural Path Planners via Slight Perturbations
Data-driven neural path planners are attracting increasing interest in the robotics community. However, their neural network components typically come as black boxes, obscuring their underlying decision-making processes. Their black-box nature exposes them to the risk of being compromised via the insertion of hidden malicious behaviors. For example, an attacker may hide behaviors that, when triggered, hijack a delivery robot by guiding it to a specific (albeit wrong) destination, trapping it in a predefined region, or inducing unnecessary energy expenditure by causing the robot to repeatedly circle a region. In this paper, we propose a novel approach to specify and inject a range of hidden malicious behaviors, known as backdoors, into neural path planners. Our approach provides a concise but flexible way to define these behaviors, and we show that hidden behaviors can be triggered by slight perturbations (e.g., inserting a tiny unnoticeable object), that can nonetheless significantly compromise their integrity. We also discuss potential techniques to identify these backdoors aimed at alleviating such risks. We demonstrate our approach on both sampling-based and search-based neural path planners.
☆ Multi-AGV Path Planning Method via Reinforcement Learning and Particle Filters
The Reinforcement Learning (RL) algorithm, renowned for its robust learning capability and search stability, has garnered significant attention and found extensive application in Automated Guided Vehicle (AGV) path planning. However, RL planning algorithms encounter challenges stemming from the substantial variance of neural networks caused by environmental instability and significant fluctuations in system structure. These challenges manifest in slow convergence speed and low learning efficiency. To tackle this issue, this paper presents the Particle Filter-Double Deep Q-Network (PF-DDQN) approach, which incorporates the Particle Filter (PF) into multi-AGV reinforcement learning path planning. The PF-DDQN method leverages the imprecise weight values of the network as state values to formulate the state space equation. Through the iterative fusion process of neural networks and particle filters, the DDQN model is optimized to acquire the optimal true weight values, thus enhancing the algorithm's efficiency. The proposed method's effectiveness and superiority are validated through numerical simulations. Overall, the simulation results demonstrate that the proposed algorithm surpasses the traditional DDQN algorithm in terms of path planning superiority and training time indicators by 92.62% and 76.88%, respectively. In conclusion, the PF-DDQN method addresses the challenges encountered by RL planning algorithms in AGV path planning. By integrating the Particle Filter and optimizing the DDQN model, the proposed method achieves enhanced efficiency and outperforms the traditional DDQN algorithm in terms of path planning superiority and training time indicators.
☆ Uncertainty-Aware Deployment of Pre-trained Language-Conditioned Imitation Learning Policies
Large-scale robotic policies trained on data from diverse tasks and robotic platforms hold great promise for enabling general-purpose robots; however, reliable generalization to new environment conditions remains a major challenge. Toward addressing this challenge, we propose a novel approach for uncertainty-aware deployment of pre-trained language-conditioned imitation learning agents. Specifically, we use temperature scaling to calibrate these models and exploit the calibrated model to make uncertainty-aware decisions by aggregating the local information of candidate actions. We implement our approach in simulation using three such pre-trained models, and showcase its potential to significantly enhance task completion rates. The accompanying code is accessible at the link: https://github.com/BobWu1998/uncertainty_quant_all.git
comment: 8 pages, 7 figures
☆ Preference-Based Planning in Stochastic Environments: From Partially-Ordered Temporal Goals to Most Preferred Policies
Human preferences are not always represented via complete linear orders: It is natural to employ partially-ordered preferences for expressing incomparable outcomes. In this work, we consider decision-making and probabilistic planning in stochastic systems modeled as Markov decision processes (MDPs), given a partially ordered preference over a set of temporally extended goals. Specifically, each temporally extended goal is expressed using a formula in Linear Temporal Logic on Finite Traces (LTL$_f$). To plan with the partially ordered preference, we introduce order theory to map a preference over temporal goals to a preference over policies for the MDP. Accordingly, a most preferred policy under a stochastic ordering induces a stochastic nondominated probability distribution over the finite paths in the MDP. To synthesize a most preferred policy, our technical approach includes two key steps. In the first step, we develop a procedure to transform a partially ordered preference over temporal goals into a computational model, called preference automaton, which is a semi-automaton with a partial order over acceptance conditions. In the second step, we prove that finding a most preferred policy is equivalent to computing a Pareto-optimal policy in a multi-objective MDP that is constructed from the original MDP, the preference automaton, and the chosen stochastic ordering relation. Throughout the paper, we employ running examples to illustrate the proposed preference specification and solution approaches. We demonstrate the efficacy of our algorithm using these examples, providing detailed analysis, and then discuss several potential future directions.
comment: arXiv admin note: substantial text overlap with arXiv:2209.12267
☆ Long and Short-Term Constraints Driven Safe Reinforcement Learning for Autonomous Driving
Reinforcement learning (RL) has been widely used in decision-making tasks, but it cannot guarantee the agent's safety in the training process due to the requirements of interaction with the environment, which seriously limits its industrial applications such as autonomous driving. Safe RL methods are developed to handle this issue by constraining the expected safety violation costs as a training objective, but they still permit unsafe state occurrence, which is unacceptable in autonomous driving tasks. Moreover, these methods are difficult to achieve a balance between the cost and return expectations, which leads to learning performance degradation for the algorithms. In this paper, we propose a novel algorithm based on the long and short-term constraints (LSTC) for safe RL. The short-term constraint aims to guarantee the short-term state safety that the vehicle explores, while the long-term constraint ensures the overall safety of the vehicle throughout the decision-making process. In addition, we develop a safe RL method with dual-constraint optimization based on the Lagrange multiplier to optimize the training process for end-to-end autonomous driving. Comprehensive experiments were conducted on the MetaDrive simulator. Experimental results demonstrate that the proposed method achieves higher safety in continuous state and action tasks, and exhibits higher exploration performance in long-distance decision-making tasks compared with state-of-the-art methods.
☆ Road Obstacle Detection based on Unknown Objectness Scores ICRA 2024
The detection of unknown traffic obstacles is vital to ensure safe autonomous driving. The standard object-detection methods cannot identify unknown objects that are not included under predefined categories. This is because object-detection methods are trained to assign a background label to pixels corresponding to the presence of unknown objects. To address this problem, the pixel-wise anomaly-detection approach has attracted increased research attention. Anomaly-detection techniques, such as uncertainty estimation and perceptual difference from reconstructed images, make it possible to identify pixels of unknown objects as out-of-distribution (OoD) samples. However, when applied to images with many unknowns and complex components, such as driving scenes, these methods often exhibit unstable performance. The purpose of this study is to achieve stable performance for detecting unknown objects by incorporating the object-detection fashions into the pixel-wise anomaly detection methods. To achieve this goal, we adopt a semantic-segmentation network with a sigmoid head that simultaneously provides pixel-wise anomaly scores and objectness scores. Our experimental results show that the objectness scores play an important role in improving the detection performance. Based on these results, we propose a novel anomaly score by integrating these two scores, which we term as unknown objectness score. Quantitative evaluations show that the proposed method outperforms state-of-the-art methods when applied to the publicly available datasets.
comment: ICRA 2024
☆ Sailing Through Point Clouds: Safe Navigation Using Point Cloud Based Control Barrier Functions
The capability to navigate safely in an unstructured environment is crucial when deploying robotic systems in real-world scenarios. Recently, control barrier function (CBF) based approaches have been highly effective in synthesizing safety-critical controllers. In this work, we propose a novel CBF-based local planner comprised of two components: Vessel and Mariner. The Vessel is a novel scaling factor based CBF formulation that synthesizes CBFs using only point cloud data. The Mariner is a CBF-based preview control framework that is used to mitigate getting stuck in spurious equilibria during navigation. To demonstrate the efficacy of our proposed approach, we first compare the proposed point cloud based CBF formulation with other point cloud based CBF formulations. Then, we demonstrate the performance of our proposed approach and its integration with global planners using experimental studies on the Unitree B1 and Unitree Go2 quadruped robots in various environments.
☆ LocoMan: Advancing Versatile Quadrupedal Dexterity with Lightweight Loco-Manipulators
Quadrupedal robots have emerged as versatile agents capable of locomoting and manipulating in complex environments. Traditional designs typically rely on the robot's inherent body parts or incorporate top-mounted arms for manipulation tasks. However, these configurations may limit the robot's operational dexterity, efficiency and adaptability, particularly in cluttered or constrained spaces. In this work, we present LocoMan, a dexterous quadrupedal robot with a novel morphology to perform versatile manipulation in diverse constrained environments. By equipping a Unitree Go1 robot with two low-cost and lightweight modular 3-DoF loco-manipulators on its front calves, LocoMan leverages the combined mobility and functionality of the legs and grippers for complex manipulation tasks that require precise 6D positioning of the end effector in a wide workspace. To harness the loco-manipulation capabilities of LocoMan, we introduce a unified control framework that extends the whole-body controller (WBC) to integrate the dynamics of loco-manipulators. Through experiments, we validate that the proposed whole-body controller can accurately and stably follow desired 6D trajectories of the end effector and torso, which, when combined with the large workspace from our design, facilitates a diverse set of challenging dexterous loco-manipulation tasks in confined spaces, such as opening doors, plugging into sockets, picking objects in narrow and low-lying spaces, and bimanual manipulation.
comment: Project page: https://linchangyi1.github.io/LocoMan
☆ SCANet: Correcting LEGO Assembly Errors with Self-Correct Assembly Network
Autonomous assembly in robotics and 3D vision presents significant challenges, particularly in ensuring assembly correctness. Presently, predominant methods such as MEPNet focus on assembling components based on manually provided images. However, these approaches often fall short in achieving satisfactory results for tasks requiring long-term planning. Concurrently, we observe that integrating a self-correction module can partially alleviate such issues. Motivated by this concern, we introduce the single-step assembly error correction task, which involves identifying and rectifying misassembled components. To support research in this area, we present the LEGO Error Correction Assembly Dataset (LEGO-ECA), comprising manual images for assembly steps and instances of assembly failures. Additionally, we propose the Self-Correct Assembly Network (SCANet), a novel method to address this task. SCANet treats assembled components as queries, determining their correctness in manual images and providing corrections when necessary. Finally, we utilize SCANet to correct the assembly results of MEPNet. Experimental results demonstrate that SCANet can identify and correct MEPNet's misassembled results, significantly improving the correctness of assembly. Our code and dataset are available at https://github.com/Yaser-wyx/SCANet.
☆ Online Embedding Multi-Scale CLIP Features into 3D Maps
This study introduces a novel approach to online embedding of multi-scale CLIP (Contrastive Language-Image Pre-Training) features into 3D maps. By harnessing CLIP, this methodology surpasses the constraints of conventional vocabulary-limited methods and enables the incorporation of semantic information into the resultant maps. While recent approaches have explored the embedding of multi-modal features in maps, they often impose significant computational costs, lacking practicality for exploring unfamiliar environments in real time. Our approach tackles these challenges by efficiently computing and embedding multi-scale CLIP features, thereby facilitating the exploration of unfamiliar environments through real-time map generation. Moreover, the embedding CLIP features into the resultant maps makes offline retrieval via linguistic queries feasible. In essence, our approach simultaneously achieves real-time object search and mapping of unfamiliar environments. Additionally, we propose a zero-shot object-goal navigation system based on our mapping approach, and we validate its efficacy through object-goal navigation, offline object retrieval, and multi-object-goal navigation in both simulated environments and real robot experiments. The findings demonstrate that our method not only exhibits swifter performance than state-of-the-art mapping methods but also surpasses them in terms of the success rate of object-goal navigation tasks.
comment: 8 pages, 7 figures
☆ Vision-Based Force Estimation for Minimally Invasive Telesurgery Through Contact Detection and Local Stiffness Models
In minimally invasive telesurgery, obtaining accurate force information is difficult due to the complexities of in-vivo end effector force sensing. This constrains development and implementation of haptic feedback and force-based automated performance metrics, respectively. Vision-based force sensing approaches using deep learning are a promising alternative to intrinsic end effector force sensing. However, they have limited ability to generalize to novel scenarios, and require learning on high-quality force sensor training data that can be difficult to obtain. To address these challenges, this paper presents a novel vision-based contact-conditional approach for force estimation in telesurgical environments. Our method leverages supervised learning with human labels and end effector position data to train deep neural networks. Predictions from these trained models are optionally combined with robot joint torque information to estimate forces indirectly from visual data. We benchmark our method against ground truth force sensor data and demonstrate generality by fine-tuning to novel surgical scenarios in a data-efficient manner. Our methods demonstrated greater than 90% accuracy on contact detection and less than 10% force prediction error. These results suggest potential usefulness of contact-conditional force estimation for sensory substitution haptic feedback and tissue handling skill evaluation in clinical settings.
comment: Preprint of an article accepted in Journal of Medical Robotics Research \copyright 2024 copyright World Scientific Publishing Company
♻ ☆ SOAC: Spatio-Temporal Overlap-Aware Multi-Sensor Calibration using Neural Radiance Fields CVPR 2024
In rapidly-evolving domains such as autonomous driving, the use of multiple sensors with different modalities is crucial to ensure high operational precision and stability. To correctly exploit the provided information by each sensor in a single common frame, it is essential for these sensors to be accurately calibrated. In this paper, we leverage the ability of Neural Radiance Fields (NeRF) to represent different sensors modalities in a common volumetric representation to achieve robust and accurate spatio-temporal sensor calibration. By designing a partitioning approach based on the visible part of the scene for each sensor, we formulate the calibration problem using only the overlapping areas. This strategy results in a more robust and accurate calibration that is less prone to failure. We demonstrate that our approach works on outdoor urban scenes by validating it on multiple established driving datasets. Results show that our method is able to get better accuracy and robustness compared to existing methods.
comment: Accepted at CVPR 2024. Project page: https://qherau.github.io/SOAC/
♻ ☆ Sim-to-Real gap in RL: Use Case with TIAGo and Isaac Sim/Gym
This paper explores policy-learning approaches in the context of sim-to-real transfer for robotic manipulation using a TIAGo mobile manipulator, focusing on two state-of-art simulators, Isaac Gym and Isaac Sim, both developed by Nvidia. Control architectures are discussed, with a particular emphasis on achieving collision-less movement in both simulation and the real environment. Presented results demonstrate successful sim-to-real transfer, showcasing similar movements executed by an RL-trained model in both simulated and real setups.
comment: Accepted in ERF24 workshop "Towards Efficient and Portable Robot Learning for Real-World Settings". To be published in Springer Proceedings in Advanced Robotics
♻ ☆ Modeling and Control of Intrinsically Elasticity Coupled Soft-Rigid Robots
While much work has been done recently in the realm of model-based control of soft robots and soft-rigid hybrids, most works examine robots that have an inherently serial structure. While these systems have been prevalent in the literature, there is an increasing trend toward designing soft-rigid hybrids with intrinsically coupled elasticity between various degrees of freedom. In this work, we seek to address the issues of modeling and controlling such structures, particularly when underactuated. We introduce several simple models for elastic coupling, typical of those seen in these systems. We then propose a controller that compensates for the elasticity, and we prove its stability with Lyapunov methods without relying on the elastic dominance assumption. This controller is applicable to the general class of underactuated soft robots. After evaluating the controller in simulated cases, we then develop a simple hardware platform to evaluate both the models and the controller. Finally, using the hardware, we demonstrate a novel use case for underactuated, elastically coupled systems in "sensorless" force control.
comment: 7 pages, 8 figures
♻ ☆ Safe Control for Soft-Rigid Robots with Self-Contact using Control Barrier Functions
Incorporating both flexible and rigid components in robot designs offers a unique solution to the limitations of traditional rigid robotics by enabling both compliance and strength. This paper explores the challenges and solutions for controlling soft-rigid hybrid robots, particularly addressing the issue of self-contact. Conventional control methods prioritize precise state tracking, inadvertently increasing the system's overall stiffness, which is not always desirable in interactions with the environment or within the robot itself. To address this, we investigate the application of Control Barrier Functions (CBFs) and High Order CBFs to manage self-contact scenarios in serially connected soft-rigid hybrid robots. Through an analysis based on Piecewise Constant Curvature (PCC) kinematics, we establish CBFs within a classical control framework for self-contact dynamics. Our methodology is rigorously evaluated in both simulation environments and physical hardware systems. The findings demonstrate that our proposed control strategy effectively regulates self-contact in soft-rigid hybrid robotic systems, marking a significant advancement in the field of robotics.
comment: 6 pages, 6 figures, submitted to IEEE Robosoft 2024 Conference
♻ ☆ DRIVE: Data-driven Robot Input Vector Exploration ICRA2024
An accurate motion model is a fundamental component of most autonomous navigation systems. While much work has been done on improving model formulation, no standard protocol exists for gathering empirical data required to train models. In this work, we address this issue by proposing Data-driven Robot Input Vector Exploration (DRIVE), a protocol that enables characterizing uncrewed ground vehicles (UGVs) input limits and gathering empirical model training data. We also propose a novel learned slip approach outperforming similar acceleration learning approaches. Our contributions are validated through an extensive experimental evaluation, cumulating over 7 km and 1.8 h of driving data over three distinct UGVs and four terrain types. We show that our protocol offers increased predictive performance over common human-driven data-gathering protocols. Furthermore, our protocol converges with 46 s of training data, almost four times less than the shortest human dataset gathering protocol. We show that the operational limit for our model is reached in extreme slip conditions encountered on surfaced ice. DRIVE is an efficient way of characterizing UGV motion in its operational conditions. Our code and dataset are both available online at this link: https://github.com/norlab-ulaval/DRIVE.
comment: 8 pages, 7 figures, 1 table, accepted for publication at the 2024 IEEE International Conference on Robotics and Automation (ICRA2024), Yokohama, Japan
♻ ☆ Frontier-Based Exploration for Multi-Robot Rendezvous in Communication-Restricted Unknown Environments
Multi-robot rendezvous and exploration are fundamental challenges in the domain of mobile robotic systems. This paper addresses multi-robot rendezvous within an initially unknown environment where communication is only possible after the rendezvous. Traditionally, exploration has been focused on rapidly mapping the environment, often leading to suboptimal rendezvous performance in later stages. We adapt a standard frontier-based exploration technique to integrate exploration and rendezvous into a unified strategy, with a mechanism that allows robots to re-visit previously explored regions thus enhancing rendezvous opportunities. We validate our approach in 3D realistic simulations using ROS, showcasing its effectiveness in achieving faster rendezvous times compared to exploration strategies.
♻ ☆ Natural-artificial hybrid swarm: Cyborg-insect group navigation in unknown obstructed soft terrain
Navigating multi-robot systems in complex terrains has always been a challenging task. This is due to the inherent limitations of traditional robots in collision avoidance, adaptation to unknown environments, and sustained energy efficiency. In order to overcome these limitations, this research proposes a solution by integrating living insects with miniature electronic controllers to enable robotic-like programmable control, and proposing a novel control algorithm for swarming. Although these creatures, called cyborg insects, have the ability to instinctively avoid collisions with neighbors and obstacles while adapting to complex terrains, there is a lack of literature on the control of multi-cyborg systems. This research gap is due to the difficulty in coordinating the movements of a cyborg system under the presence of insects' inherent individual variability in their reactions to control input. In response to this issue, we propose a novel swarm navigation algorithm addressing these challenges. The effectiveness of the algorithm is demonstrated through an experimental validation in which a cyborg swarm was successfully navigated through an unknown sandy field with obstacles and hills. This research contributes to the domain of swarm robotics and showcases the potential of integrating biological organisms with robotics and control theory to create more intelligent autonomous systems with real-world applications.
♻ ☆ Polygonal Cone Control Barrier Functions (PolyC2BF) for safe navigation in cluttered environments
In fields such as mining, search and rescue, and archaeological exploration, ensuring real-time, collision-free navigation of robots in confined, cluttered environments is imperative. Despite the value of established path planning algorithms, they often face challenges in convergence rates and handling dynamic infeasibilities. Alternative techniques like collision cones struggle to accurately represent complex obstacle geometries. This paper introduces a novel category of control barrier functions, known as Polygonal Cone Control Barrier Function (PolyC2BF), which addresses overestimation and computational complexity issues. The proposed PolyC2BF, formulated as a Quadratic Programming (QP) problem, proves effective in facilitating collision-free movement of multiple robots in complex environments. The efficacy of this approach is further demonstrated through PyBullet simulations on quadruped (unicycle model), and crazyflie 2.1 (quadrotor model) in cluttered environments.
comment: 6 Pages, 6 Figures. Accepted at European Control Conference (ECC) 2024. arXiv admin note: text overlap with arXiv:2303.15871
♻ ☆ Risk-aware Control for Robots with Non-Gaussian Belief Spaces
This paper addresses the problem of safety-critical control of autonomous robots, considering the ubiquitous uncertainties arising from unmodeled dynamics and noisy sensors. To take into account these uncertainties, probabilistic state estimators are often deployed to obtain a belief over possible states. Namely, Particle Filters (PFs) can handle arbitrary non-Gaussian distributions in the robot's state. In this work, we define the belief state and belief dynamics for continuous-discrete PFs and construct safe sets in the underlying belief space. We design a controller that provably keeps the robot's belief state within this safe set. As a result, we ensure that the risk of the unknown robot's state violating a safety specification, such as avoiding a dangerous area, is bounded. We provide an open-source implementation as a ROS2 package and evaluate the solution in simulations and hardware experiments involving high-dimensional belief spaces.
Learning Quadruped Locomotion Using Differentiable Simulation
While most recent advancements in legged robot control have been driven by model-free reinforcement learning, we explore the potential of differentiable simulation. Differentiable simulation promises faster convergence and more stable training by computing low-variant first-order gradients using the robot model, but so far, its use for legged robot control has remained limited to simulation. The main challenge with differentiable simulation lies in the complex optimization landscape of robotic tasks due to discontinuities in contact-rich environments, e.g., quadruped locomotion. This work proposes a new, differentiable simulation framework to overcome these challenges. The key idea involves decoupling the complex whole-body simulation, which may exhibit discontinuities due to contact, into two separate continuous domains. Subsequently, we align the robot state resulting from the simplified model with a more precise, non-differentiable simulator to maintain sufficient simulation accuracy. Our framework enables learning quadruped walking in minutes using a single simulated robot without any parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills, including trot, pace, bound, and gallop, on challenging terrains in minutes. Additionally, our policy achieves robust locomotion performance in the real world zero-shot. To the best of our knowledge, this work represents the first demonstration of using differentiable simulation for controlling a real quadruped robot. This work provides several important insights into using differentiable simulations for legged locomotion in the real world.
♻ ☆ Non-smooth Control Barrier Functions for Stochastic Dynamical Systems
Uncertainties arising in various control systems, such as robots that are subject to unknown disturbances or environmental variations, pose significant challenges for ensuring system safety, such as collision avoidance. At the same time, safety specifications are getting more and more complex, e.g., by composing multiple safety objectives through Boolean operators resulting in non-smooth descriptions of safe sets. Control Barrier Functions (CBFs) have emerged as a control technique to provably guarantee system safety. In most settings, they rely on an assumption of having deterministic dynamics and smooth safe sets. This paper relaxes these two assumptions by extending CBFs to encompass control systems with stochastic dynamics and safe sets defined by non-smooth functions. By explicitly considering the stochastic nature of system dynamics and accommodating complex safety specifications, our method enables the design of safe control strategies in uncertain and complex systems. We provide formal guarantees on the safety of the system by leveraging the theoretical foundations of stochastic CBFs and non-smooth safe sets. Numerical simulations demonstrate the effectiveness of the approach in various scenarios.
♻ ☆ MMP++: Motion Manifold Primitives with Parametric Curve Models
Motion Manifold Primitives (MMP), a manifold-based approach for encoding basic motion skills, can produce diverse trajectories, enabling the system to adapt to unseen constraints. Nonetheless, we argue that current MMP models lack crucial functionalities of movement primitives, such as temporal and via-points modulation, found in traditional approaches. This shortfall primarily stems from MMP's reliance on discrete-time trajectories. To overcome these limitations, we introduce Motion Manifold Primitives++ (MMP++), a new model that integrates the strengths of both MMP and traditional methods by incorporating parametric curve representations into the MMP framework. Furthermore, we identify a significant challenge with MMP++: performance degradation due to geometric distortions in the latent space, meaning that similar motions are not closely positioned. To address this, Isometric Motion Manifold Primitives++ (IMMP++) is proposed to ensure the latent space accurately preserves the manifold's geometry. Our experimental results across various applications, including 2-DoF planar motions, 7-DoF robot arm motions, and SE(3) trajectory planning, show that MMP++ and IMMP++ outperform existing methods in trajectory generation tasks, achieving substantial improvements in some cases. Moreover, they enable the modulation of latent coordinates and via-points, thereby allowing efficient online adaptation to dynamic environments.
comment: 12 pages. This work has been submitted to the IEEE for possible publication
♻ ☆ RoboDuet: A Framework Affording Mobile-Manipulation and Cross-Embodiment
Combining the mobility of legged robots with the manipulation skills of arms has the potential to significantly expand the operational range and enhance the capabilities of robotic systems in performing various mobile manipulation tasks. Existing approaches are confined to imprecise six degrees of freedom (DoF) manipulation and possess a limited arm workspace. In this paper, we propose a novel framework, RoboDuet, which employs two collaborative policies to realize locomotion and manipulation simultaneously, achieving whole-body control through interactions between each other. Surprisingly, going beyond the large-range pose tracking, we find that the two-policy framework may enable cross-embodiment deployment such as using different quadrupedal robots or other arms. Our experiments demonstrate that the policies trained through RoboDuet can accomplish stable gaits, agile 6D end-effector pose tracking, and zero-shot exchange of legged robots, and can be deployed in the real world to perform various mobile manipulation tasks. Our project page with demo videos is at https://locomanip-duet.github.io .
♻ ☆ Nigel -- Mechatronic Design and Robust Sim2Real Control of an Over-Actuated Autonomous Vehicle
Simulation to reality (sim2real) transfer from a dynamics and controls perspective usually involves re-tuning or adapting the designed algorithms to suit real-world operating conditions, which often violates the performance guarantees established originally. This work presents a generalizable framework for achieving reliable sim2real transfer of autonomy-oriented control systems using multi-model multi-objective robust optimal control synthesis, which lends well to uncertainty handling and disturbance rejection with theoretical guarantees. Particularly, this work is centered around a novel actuation-redundant scaled autonomous vehicle called Nigel, with independent all-wheel drive and independent all-wheel steering architecture, whose enhanced configuration space bodes well for robust control applications. To this end, we present the mechatronic design, dynamics modeling, parameter identification, and robust stabilizing as well as tracking control of Nigel using the proposed framework, with exhaustive experimentation and benchmarking in simulation as well as real-world settings.
♻ ☆ PPAD: Iterative Interactions of Prediction and Planning for End-to-end Autonomous Driving
We present a new interaction mechanism of prediction and planning for end-to-end autonomous driving, called PPAD (Iterative Interaction of Prediction and Planning Autonomous Driving), which considers the timestep-wise interaction to better integrate prediction and planning. An ego vehicle performs motion planning at each timestep based on the trajectory prediction of surrounding agents (e.g., vehicles and pedestrians) and its local road conditions. Unlike existing end-to-end autonomous driving frameworks, PPAD models the interactions among ego, agents, and the dynamic environment in an autoregressive manner by interleaving the Prediction and Planning processes at every timestep, instead of a single sequential process of prediction followed by planning. Specifically, we design ego-to-agent, ego-to-map, and ego-to-BEV interaction mechanisms with hierarchical dynamic key objects attention to better model the interactions. The experiments on the nuScenes benchmark show that our approach outperforms state-of-the-art methods.
♻ ☆ SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System
Accuracy and computational efficiency are the most important metrics to Visual Inertial Navigation System (VINS). The existing VINS algorithms with either high accuracy or low computational complexity, are difficult to provide the high precision localization in resource-constrained devices. To this end, we propose a novel filter-based VINS framework named SchurVINS, which could guarantee both high accuracy by building a complete residual model and low computational complexity with Schur complement. Technically, we first formulate the full residual model where Gradient, Hessian and observation covariance are explicitly modeled. Then Schur complement is employed to decompose the full model into ego-motion residual model and landmark residual model. Finally, Extended Kalman Filter (EKF) update is implemented in these two models with high efficiency. Experiments on EuRoC and TUM-VI datasets show that our method notably outperforms state-of-the-art (SOTA) methods in both accuracy and computational complexity. The experimental code of SchurVINS is available at https://github.com/bytedance/SchurVINS.
♻ ☆ DiffPrompter: Differentiable Implicit Visual Prompts for Semantic-Segmentation in Adverse Conditions
Semantic segmentation in adverse weather scenarios is a critical task for autonomous driving systems. While foundation models have shown promise, the need for specialized adaptors becomes evident for handling more challenging scenarios. We introduce DiffPrompter, a novel differentiable visual and latent prompting mechanism aimed at expanding the learning capabilities of existing adaptors in foundation models. Our proposed $\nabla$HFC image processing block excels particularly in adverse weather conditions, where conventional methods often fall short. Furthermore, we investigate the advantages of jointly training visual and latent prompts, demonstrating that this combined approach significantly enhances performance in out-of-distribution scenarios. Our differentiable visual prompts leverage parallel and series architectures to generate prompts, effectively improving object segmentation tasks in adverse conditions. Through a comprehensive series of experiments and evaluations, we provide empirical evidence to support the efficacy of our approach. Project page at https://diffprompter.github.io.
Leveraging Symmetry in RL-based Legged Locomotion Control
Model-free reinforcement learning is a promising approach for autonomously solving challenging robotics control problems, but faces exploration difficulty without information of the robot's kinematics and dynamics morphology. The under-exploration of multiple modalities with symmetric states leads to behaviors that are often unnatural and sub-optimal. This issue becomes particularly pronounced in the context of robotic systems with morphological symmetries, such as legged robots for which the resulting asymmetric and aperiodic behaviors compromise performance, robustness, and transferability to real hardware. To mitigate this challenge, we can leverage symmetry to guide and improve the exploration in policy learning via equivariance/invariance constraints. In this paper, we investigate the efficacy of two approaches to incorporate symmetry: modifying the network architectures to be strictly equivariant/invariant, and leveraging data augmentation to approximate equivariant/invariant actor-critics. We implement the methods on challenging loco-manipulation and bipedal locomotion tasks and compare with an unconstrained baseline. We find that the strictly equivariant policy consistently outperforms other methods in sample efficiency and task performance in simulation. In addition, symmetry-incorporated approaches exhibit better gait quality, higher robustness and can be deployed zero-shot in real-world experiments.
♻ ☆ Optimal Sensor Deception to Deviate from an Allowed Itinerary
In this work, we study a class of deception planning problems in which an agent aims to alter a security monitoring system's sensor readings so as to disguise its adversarial itinerary as an allowed itinerary in the environment. The adversarial itinerary set and allowed itinerary set are captured by regular languages. To deviate without being detected, we investigate whether there exists a strategy for the agent to alter the sensor readings, with a minimal cost, such that for any of those paths it takes, the system thinks the agent took a path within the allowed itinerary. Our formulation assumes an offline sensor alteration where the agent determines the sensor alteration strategy and implement it, and then carry out any path in its deviation itinerary. We prove that the problem of solving the optimal sensor alteration is NP-hard, by a reduction from the directed multi-cut problem. Further, we present an exact algorithm based on integer linear programming and demonstrate the correctness and the efficacy of the algorithm in case studies.
Artificial Intelligence 139
☆ Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models
In this work, we introduce Mini-Gemini, a simple and effective framework enhancing multi-modality Vision Language Models (VLMs). Despite the advancements in VLMs facilitating basic visual dialog and reasoning, a performance gap persists compared to advanced models like GPT-4 and Gemini. We try to narrow the gap by mining the potential of VLMs for better performance and any-to-any workflow from three aspects, i.e., high-resolution visual tokens, high-quality data, and VLM-guided generation. To enhance visual tokens, we propose to utilize an additional visual encoder for high-resolution refinement without increasing the visual token count. We further construct a high-quality dataset that promotes precise image comprehension and reasoning-based generation, expanding the operational scope of current VLMs. In general, Mini-Gemini further mines the potential of VLMs and empowers current frameworks with image understanding, reasoning, and generation simultaneously. Mini-Gemini supports a series of dense and MoE Large Language Models (LLMs) from 2B to 34B. It is demonstrated to achieve leading performance in several zero-shot benchmarks and even surpasses the developed private models. Code and models are available at https://github.com/dvlab-research/MiniGemini.
comment: Code and models are available at https://github.com/dvlab-research/MiniGemini
☆ ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation CVPR
In the absence of parallax cues, a learning-based single image depth estimation (SIDE) model relies heavily on shading and contextual cues in the image. While this simplicity is attractive, it is necessary to train such models on large and varied datasets, which are difficult to capture. It has been shown that using embeddings from pre-trained foundational models, such as CLIP, improves zero shot transfer in several applications. Taking inspiration from this, in our paper we explore the use of global image priors generated from a pre-trained ViT model to provide more detailed contextual information. We argue that the embedding vector from a ViT model, pre-trained on a large dataset, captures greater relevant information for SIDE than the usual route of generating pseudo image captions, followed by CLIP based text embeddings. Based on this idea, we propose a new SIDE model using a diffusion backbone which is conditioned on ViT embeddings. Our proposed design establishes a new state-of-the-art (SOTA) for SIDE on NYUv2 dataset, achieving Abs Rel error of 0.059(14% improvement) compared to 0.069 by the current SOTA (VPD). And on KITTI dataset, achieving Sq Rel error of 0.139 (2% improvement) compared to 0.142 by the current SOTA (GEDepth). For zero-shot transfer with a model trained on NYUv2, we report mean relative improvement of (20%, 23%, 81%, 25%) over NeWCRFs on (Sun-RGBD, iBims1, DIODE, HyperSim) datasets, compared to (16%, 18%, 45%, 9%) by ZoeDepth. The code is available at https://github.com/Aradhye2002/EcoDepth.
comment: Accepted at IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024
☆ Long-form factuality in large language models
Large language models (LLMs) often generate content that contains factual errors when responding to fact-seeking prompts on open-ended topics. To benchmark a model's long-form factuality in open domains, we first use GPT-4 to generate LongFact, a prompt set comprising thousands of questions spanning 38 topics. We then propose that LLM agents can be used as automated evaluators for long-form factuality through a method which we call Search-Augmented Factuality Evaluator (SAFE). SAFE utilizes an LLM to break down a long-form response into a set of individual facts and to evaluate the accuracy of each fact using a multi-step reasoning process comprising sending search queries to Google Search and determining whether a fact is supported by the search results. Furthermore, we propose extending F1 score as an aggregated metric for long-form factuality. To do so, we balance the percentage of supported facts in a response (precision) with the percentage of provided facts relative to a hyperparameter representing a user's preferred response length (recall). Empirically, we demonstrate that LLM agents can achieve superhuman rating performance - on a set of ~16k individual facts, SAFE agrees with crowdsourced human annotators 72% of the time, and on a random subset of 100 disagreement cases, SAFE wins 76% of the time. At the same time, SAFE is more than 20 times cheaper than human annotators. We also benchmark thirteen language models on LongFact across four model families (Gemini, GPT, Claude, and PaLM-2), finding that larger language models generally achieve better long-form factuality. LongFact, SAFE, and all experimental code are available at https://github.com/google-deepmind/long-form-factuality.
☆ Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction
We tackle the challenge of efficiently reconstructing a 3D asset from a single image with growing demands for automated 3D content creation pipelines. Previous methods primarily rely on Score Distillation Sampling (SDS) and Neural Radiance Fields (NeRF). Despite their significant success, these approaches encounter practical limitations due to lengthy optimization and considerable memory usage. In this report, we introduce Gamba, an end-to-end amortized 3D reconstruction model from single-view images, emphasizing two main insights: (1) 3D representation: leveraging a large number of 3D Gaussians for an efficient 3D Gaussian splatting process; (2) Backbone design: introducing a Mamba-based sequential network that facilitates context-dependent reasoning and linear scalability with the sequence (token) length, accommodating a substantial number of Gaussians. Gamba incorporates significant advancements in data preprocessing, regularization design, and training methodologies. We assessed Gamba against existing optimization-based and feed-forward 3D generation approaches using the real-world scanned OmniObject3D dataset. Here, Gamba demonstrates competitive generation capabilities, both qualitatively and quantitatively, while achieving remarkable speed, approximately 0.6 second on a single NVIDIA A100 GPU.
☆ ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object CVPR 2024
We establish rigorous benchmarks for visual perception robustness. Synthetic images such as ImageNet-C, ImageNet-9, and Stylized ImageNet provide specific type of evaluation over synthetic corruptions, backgrounds, and textures, yet those robustness benchmarks are restricted in specified variations and have low synthetic quality. In this work, we introduce generative model as a data source for synthesizing hard images that benchmark deep models' robustness. Leveraging diffusion models, we are able to generate images with more diversified backgrounds, textures, and materials than any prior work, where we term this benchmark as ImageNet-D. Experimental results show that ImageNet-D results in a significant accuracy drop to a range of vision models, from the standard ResNet visual classifier to the latest foundation models like CLIP and MiniGPT-4, significantly reducing their accuracy by up to 60\%. Our work suggests that diffusion models can be an effective source to test vision models. The code and dataset are available at https://github.com/chenshuang-zhang/imagenet_d.
comment: Accepted at CVPR 2024
☆ Superior Parallel Big Data Clustering through Competitive Stochastic Sample Size Optimization in Big-means
This paper introduces a novel K-means clustering algorithm, an advancement on the conventional Big-means methodology. The proposed method efficiently integrates parallel processing, stochastic sampling, and competitive optimization to create a scalable variant designed for big data applications. It addresses scalability and computation time challenges typically faced with traditional techniques. The algorithm adjusts sample sizes dynamically for each worker during execution, optimizing performance. Data from these sample sizes are continually analyzed, facilitating the identification of the most efficient configuration. By incorporating a competitive element among workers using different sample sizes, efficiency within the Big-means algorithm is further stimulated. In essence, the algorithm balances computational time and clustering quality by employing a stochastic, competitive sampling strategy in a parallel computing setting.
☆ ModaLink: Unifying Modalities for Efficient Image-to-PointCloud Place Recognition
Place recognition is an important task for robots and autonomous cars to localize themselves and close loops in pre-built maps. While single-modal sensor-based methods have shown satisfactory performance, cross-modal place recognition that retrieving images from a point-cloud database remains a challenging problem. Current cross-modal methods transform images into 3D points using depth estimation for modality conversion, which are usually computationally intensive and need expensive labeled data for depth supervision. In this work, we introduce a fast and lightweight framework to encode images and point clouds into place-distinctive descriptors. We propose an effective Field of View (FoV) transformation module to convert point clouds into an analogous modality as images. This module eliminates the necessity for depth estimation and helps subsequent modules achieve real-time performance. We further design a non-negative factorization-based encoder to extract mutually consistent semantic features between point clouds and images. This encoder yields more distinctive global descriptors for retrieval. Experimental results on the KITTI dataset show that our proposed methods achieve state-of-the-art performance while running in real time. Additional evaluation on the HAOMO dataset covering a 17 km trajectory further shows the practical generalization capabilities. We have released the implementation of our methods as open source at: https://github.com/haomo-ai/ModaLink.git.
comment: 8 pages, 11 figures, conference
☆ Detection of subclinical atherosclerosis by image-based deep learning on chest x-ray
Aims. To develop a deep-learning based system for recognition of subclinical atherosclerosis on a plain frontal chest x-ray. Methods and Results. A deep-learning algorithm to predict coronary artery calcium (CAC) score (the AI-CAC model) was developed on 460 chest x-ray (80% training cohort, 20% internal validation cohort) of primary prevention patients (58.4% male, median age 63 [51-74] years) with available paired chest x-ray and chest computed tomography (CT) indicated for any clinical reason and performed within 3 months. The CAC score calculated on chest CT was used as ground truth. The model was validated on an temporally-independent cohort of 90 patients from the same institution (external validation). The diagnostic accuracy of the AI-CAC model assessed by the area under the curve (AUC) was the primary outcome. Overall, median AI-CAC score was 35 (0-388) and 28.9% patients had no AI-CAC. AUC of the AI-CAC model to identify a CAC>0 was 0.90 in the internal validation cohort and 0.77 in the external validation cohort. Sensitivity was consistently above 92% in both cohorts. In the overall cohort (n=540), among patients with AI-CAC=0, a single ASCVD event occurred, after 4.3 years. Patients with AI-CAC>0 had significantly higher Kaplan Meier estimates for ASCVD events (13.5% vs. 3.4%, log-rank=0.013). Conclusion. The AI-CAC model seems to accurately detect subclinical atherosclerosis on chest x-ray with elevated sensitivity, and to predict ASCVD events with elevated negative predictive value. Adoption of the AI-CAC model to refine CV risk stratification or as an opportunistic screening tool requires prospective evaluation.
comment: Submitted to European Heart Journal - Cardiovascular Imaging Added also the additional material 44 pages (30 main paper, 14 additional material), 14 figures (5 main manuscript, 9 additional material)
☆ Many-Objective Evolutionary Influence Maximization: Balancing Spread, Budget, Fairness, and Time GECCO 24
The Influence Maximization (IM) problem seeks to discover the set of nodes in a graph that can spread the information propagation at most. This problem is known to be NP-hard, and it is usually studied by maximizing the influence (spread) and, optionally, optimizing a second objective, such as minimizing the seed set size or maximizing the influence fairness. However, in many practical scenarios multiple aspects of the IM problem must be optimized at the same time. In this work, we propose a first case study where several IM-specific objective functions, namely budget, fairness, communities, and time, are optimized on top of the maximization of influence and minimization of the seed set size. To this aim, we introduce MOEIM (Many-Objective Evolutionary Algorithm for Influence Maximization) a Multi-Objective Evolutionary Algorithm (MOEA) based on NSGA-II incorporating graph-aware operators and a smart initialization. We compare MOEIM in two experimental settings, including a total of nine graph datasets, two heuristic methods, a related MOEA, and a state-of-the-art Deep Learning approach. The experiments show that MOEIM overall outperforms the competitors in most of the tested many-objective settings. To conclude, we also investigate the correlation between the objectives, leading to novel insights into the topic. The codebase is available at https://github.com/eliacunegatti/MOEIM.
comment: To appear in Genetic and Evolutionary Computation Conference (GECCO 24 Companion), July 14 18, 2024, Melbourne, VIC, Australia. ACM, New York, NY, USA
☆ Understanding the Learning Dynamics of Alignment with Human Feedback
Aligning large language models (LLMs) with human intentions has become a critical task for safely deploying models in real-world systems. While existing alignment approaches have seen empirical success, theoretically understanding how these methods affect model behavior remains an open question. Our work provides an initial attempt to theoretically analyze the learning dynamics of human preference alignment. We formally show how the distribution of preference datasets influences the rate of model updates and provide rigorous guarantees on the training accuracy. Our theory also reveals an intricate phenomenon where the optimization is prone to prioritizing certain behaviors with higher preference distinguishability. We empirically validate our findings on contemporary LLMs and alignment tasks, reinforcing our theoretical insights and shedding light on considerations for future alignment approaches. Disclaimer: This paper contains potentially offensive text; reader discretion is advised.
☆ Enhancing Manufacturing Quality Prediction Models through the Integration of Explainability Methods
This research presents a method that utilizes explainability techniques to amplify the performance of machine learning (ML) models in forecasting the quality of milling processes, as demonstrated in this paper through a manufacturing use case. The methodology entails the initial training of ML models, followed by a fine-tuning phase where irrelevant features identified through explainability methods are eliminated. This procedural refinement results in performance enhancements, paving the way for potential reductions in manufacturing costs and a better understanding of the trained ML models. This study highlights the usefulness of explainability techniques in both explaining and optimizing predictive models in the manufacturing realm.
☆ Probabilistic Model Checking of Stochastic Reinforcement Learning Policies
We introduce a method to verify stochastic reinforcement learning (RL) policies. This approach is compatible with any RL algorithm as long as the algorithm and its corresponding environment collectively adhere to the Markov property. In this setting, the future state of the environment should depend solely on its current state and the action executed, independent of any previous states or actions. Our method integrates a verification technique, referred to as model checking, with RL, leveraging a Markov decision process, a trained RL policy, and a probabilistic computation tree logic (PCTL) formula to build a formal model that can be subsequently verified via the model checker Storm. We demonstrate our method's applicability across multiple benchmarks, comparing it to baseline methods called deterministic safety estimates and naive monolithic model checking. Our results show that our method is suited to verify stochastic RL policies.
☆ Semi-Supervised Learning for Deep Causal Generative Models
Developing models that can answer questions of the form "How would $x$ change if $y$ had been $z$?" is fundamental for advancing medical image analysis. Training causal generative models that address such counterfactual questions, though, currently requires that all relevant variables have been observed and that corresponding labels are available in training data. However, clinical data may not have complete records for all patients and state of the art causal generative models are unable to take full advantage of this. We thus develop, for the first time, a semi-supervised deep causal generative model that exploits the causal relationships between variables to maximise the use of all available data. We explore this in the setting where each sample is either fully labelled or fully unlabelled, as well as the more clinically realistic case of having different labels missing for each sample. We leverage techniques from causal inference to infer missing values and subsequently generate realistic counterfactuals, even for samples with incomplete labels.
☆ Mitigating Hallucinations in Large Vision-Language Models with Instruction Contrastive Decoding
Large Vision-Language Models (LVLMs) are increasingly adept at generating contextually detailed and coherent responses from visual inputs. However, their application in multimodal decision-making and open-ended generation is hindered by a notable rate of hallucinations, where generated text inaccurately represents the visual contents. To address this issue, this paper introduces the Instruction Contrastive Decoding (ICD) method, a novel approach designed to reduce hallucinations during LVLM inference. Our method is inspired by our observation that what we call disturbance instructions significantly exacerbate hallucinations in multimodal fusion modules. ICD contrasts distributions from standard and instruction disturbance, thereby increasing alignment uncertainty and effectively subtracting hallucinated concepts from the original distribution. Through comprehensive experiments on discriminative benchmarks (POPE and MME) and a generative benchmark (LLaVa-Bench), we demonstrate that ICD significantly mitigates both object-level and attribute-level hallucinations. Moreover, our method not only addresses hallucinations but also significantly enhances the general perception and recognition capabilities of LVLMs.
☆ SAT-NGP : Unleashing Neural Graphics Primitives for Fast Relightable Transient-Free 3D reconstruction from Satellite Imagery
Current stereo-vision pipelines produce high accuracy 3D reconstruction when using multiple pairs or triplets of satellite images. However, these pipelines are sensitive to the changes between images that can occur as a result of multi-date acquisitions. Such variations are mainly due to variable shadows, reflexions and transient objects (cars, vegetation). To take such changes into account, Neural Radiance Fields (NeRF) have recently been applied to multi-date satellite imagery. However, Neural methods are very compute-intensive, taking dozens of hours to learn, compared with minutes for standard stereo-vision pipelines. Following the ideas of Instant Neural Graphics Primitives we propose to use an efficient sampling strategy and multi-resolution hash encoding to accelerate the learning. Our model, Satellite Neural Graphics Primitives (SAT-NGP) decreases the learning time to 15 minutes while maintaining the quality of the 3D reconstruction.
comment: 5 pages, 3 figures, 1 table; Accepted to International Geoscience and Remote Sensing Symposium (IGARSS) 2024; Code available at https://github.com/Ellimac0/SAT-NGP
☆ Contrastive Learning with Orthonormal Anchors (CLOA)
This study focuses on addressing the instability issues prevalent in contrastive learning, specifically examining the InfoNCE loss function and its derivatives. We reveal a critical observation that these loss functions exhibit a restrictive behavior, leading to a convergence phenomenon where embeddings tend to merge into a singular point. This "over-fusion" effect detrimentally affects classification accuracy in subsequent supervised-learning tasks. Through theoretical analysis, we demonstrate that embeddings, when equalized or confined to a rank-1 linear subspace, represent a local minimum for InfoNCE. In response to this challenge, our research introduces an innovative strategy that leverages the same or fewer labeled data than typically used in the fine-tuning phase. The loss we proposed, Orthonormal Anchor Regression Loss, is designed to disentangle embedding clusters, significantly enhancing the distinctiveness of each embedding while simultaneously ensuring their aggregation into dense, well-defined clusters. Our method demonstrates remarkable improvements with just a fraction of the conventional label requirements, as evidenced by our results on CIFAR10 and CIFAR100 datasets.
comment: 11 pages, 4 figures
☆ Annolid: Annotate, Segment, and Track Anything You Need
Annolid is a deep learning-based software package designed for the segmentation, labeling, and tracking of research targets within video files, focusing primarily on animal behavior analysis. Based on state-of-the-art instance segmentation methods, Annolid now harnesses the Cutie video object segmentation model to achieve resilient, markerless tracking of multiple animals from single annotated frames, even in environments in which they may be partially or entirely concealed by environmental features or by one another. Our integration of Segment Anything and Grounding-DINO strategies additionally enables the automatic masking and segmentation of recognizable animals and objects by text command, removing the need for manual annotation. Annolid's comprehensive approach to object segmentation flexibly accommodates a broad spectrum of behavior analysis applications, enabling the classification of diverse behavioral states such as freezing, digging, pup huddling, and social interactions in addition to the tracking of animals and their body parts.
☆ TransFusion: Contrastive Learning with Transformers
This paper proposes a novel framework, TransFusion, designed to make the process of contrastive learning more analytical and explainable. TransFusion consists of attention blocks whose softmax being replaced by ReLU, and its final block's weighted-sum operation is truncated to leave the adjacency matrix as the output. The model is trained by minimizing the Jensen-Shannon Divergence between its output and the target affinity matrix, which indicates whether each pair of samples belongs to the same or different classes. The main contribution of TransFusion lies in defining a theoretical limit for answering two fundamental questions in the field: the maximum level of data augmentation and the minimum batch size required for effective contrastive learning. Furthermore, experimental results indicate that TransFusion successfully extracts features that isolate clusters from complex real-world data, leading to improved classification accuracy in downstream tasks.
comment: 17 pages, 4 figures,
☆ Aiming for Relevance
Vital signs are crucial in intensive care units (ICUs). They are used to track the patient's state and to identify clinically significant changes. Predicting vital sign trajectories is valuable for early detection of adverse events. However, conventional machine learning metrics like RMSE often fail to capture the true clinical relevance of such predictions. We introduce novel vital sign prediction performance metrics that align with clinical contexts, focusing on deviations from clinical norms, overall trends, and trend deviations. These metrics are derived from empirical utility curves obtained in a previous study through interviews with ICU clinicians. We validate the metrics' usefulness using simulated and real clinical datasets (MIMIC and eICU). Furthermore, we employ these metrics as loss functions for neural networks, resulting in models that excel in predicting clinically significant events. This research paves the way for clinically relevant machine learning model evaluation and optimization, promising to improve ICU patient care. 10 pages, 9 figures.
comment: 10 pages, 9 figures, AMIA Informatics 2024
☆ INEXA: Interactive and Explainable Process Model Abstraction Through Object-Centric Process Mining
Process events are recorded by multiple information systems at different granularity levels. Based on the resulting event logs, process models are discovered at different granularity levels, as well. Events stored at a fine-grained granularity level, for example, may hinder the discovered process model to be displayed due the high number of resulting model elements. The discovered process model of a real-world manufacturing process, for example, consists of 1,489 model elements and over 2,000 arcs. Existing process model abstraction techniques could help reducing the size of the model, but would disconnect it from the underlying event log. Existing event abstraction techniques do neither support the analysis of mixed granularity levels, nor interactive exploration of a suitable granularity level. To enable the exploration of discovered process models at different granularity levels, we propose INEXA, an interactive, explainable process model abstraction method that keeps the link to the event log. As a starting point, INEXA aggregates large process models to a "displayable" size, e.g., for the manufacturing use case to a process model with 58 model elements. Then, the process analyst can explore granularity levels interactively, while applied abstractions are automatically traced in the event log for explainability.
☆ Spikewhisper: Temporal Spike Backdoor Attacks on Federated Neuromorphic Learning over Low-power Devices
Federated neuromorphic learning (FedNL) leverages event-driven spiking neural networks and federated learning frameworks to effectively execute intelligent analysis tasks over amounts of distributed low-power devices but also perform vulnerability to poisoning attacks. The threat of backdoor attacks on traditional deep neural networks typically comes from time-invariant data. However, in FedNL, unknown threats may be hidden in time-varying spike signals. In this paper, we start to explore a novel vulnerability of FedNL-based systems with the concept of time division multiplexing, termed Spikewhisper, which allows attackers to evade detection as much as possible, as multiple malicious clients can imperceptibly poison with different triggers at different timeslices. In particular, the stealthiness of Spikewhisper is derived from the time-domain divisibility of global triggers, in which each malicious client pastes only one local trigger to a certain timeslice in the neuromorphic sample, and also the polarity and motion of each local trigger can be configured by attackers. Extensive experiments based on two different neuromorphic datasets demonstrate that the attack success rate of Spikewispher is higher than the temporally centralized attacks. Besides, it is validated that the effect of Spikewispher is sensitive to the trigger duration.
☆ RAP: Retrieval-Augmented Planner for Adaptive Procedure Planning in Instructional Videos
Procedure Planning in instructional videos entails generating a sequence of action steps based on visual observations of the initial and target states. Despite the rapid progress in this task, there remain several critical challenges to be solved: (1) Adaptive procedures: Prior works hold an unrealistic assumption that the number of action steps is known and fixed, leading to non-generalizable models in real-world scenarios where the sequence length varies. (2) Temporal relation: Understanding the step temporal relation knowledge is essential in producing reasonable and executable plans. (3) Annotation cost: Annotating instructional videos with step-level labels (i.e., timestamp) or sequence-level labels (i.e., action category) is demanding and labor-intensive, limiting its generalizability to large-scale datasets.In this work, we propose a new and practical setting, called adaptive procedure planning in instructional videos, where the procedure length is not fixed or pre-determined. To address these challenges we introduce Retrieval-Augmented Planner (RAP) model. Specifically, for adaptive procedures, RAP adaptively determines the conclusion of actions using an auto-regressive model architecture. For temporal relation, RAP establishes an external memory module to explicitly retrieve the most relevant state-action pairs from the training videos and revises the generated procedures. To tackle high annotation cost, RAP utilizes a weakly-supervised learning manner to expand the training dataset to other task-relevant, unannotated videos by generating pseudo labels for action steps. Experiments on CrossTask and COIN benchmarks show the superiority of RAP over traditional fixed-length models, establishing it as a strong baseline solution for adaptive procedure planning.
comment: 23 pages, 6 figures, 12 tables
☆ Homogeneous Tokenizer Matters: Homogeneous Visual Tokenizer for Remote Sensing Image Understanding
The tokenizer, as one of the fundamental components of large models, has long been overlooked or even misunderstood in visual tasks. One key factor of the great comprehension power of the large language model is that natural language tokenizers utilize meaningful words or subwords as the basic elements of language. In contrast, mainstream visual tokenizers, represented by patch-based methods such as Patch Embed, rely on meaningless rectangular patches as basic elements of vision, which cannot serve as effectively as words or subwords in language. Starting from the essence of the tokenizer, we defined semantically independent regions (SIRs) for vision. We designed a simple HOmogeneous visual tOKenizer: HOOK. HOOK mainly consists of two modules: the Object Perception Module (OPM) and the Object Vectorization Module (OVM). To achieve homogeneity, the OPM splits the image into 4*4 pixel seeds and then utilizes the attention mechanism to perceive SIRs. The OVM employs cross-attention to merge seeds within the same SIR. To achieve adaptability, the OVM defines a variable number of learnable vectors as cross-attention queries, allowing for the adjustment of token quantity. We conducted experiments on the NWPU-RESISC45, WHU-RS19 classification dataset, and GID5 segmentation dataset for sparse and dense tasks. The results demonstrate that the visual tokens obtained by HOOK correspond to individual objects, which demonstrates homogeneity. HOOK outperformed Patch Embed by 6\% and 10\% in the two tasks and achieved state-of-the-art performance compared to the baselines used for comparison. Compared to Patch Embed, which requires more than one hundred tokens for one image, HOOK requires only 6 and 8 tokens for sparse and dense tasks, respectively, resulting in efficiency improvements of 1.5 to 2.8 times. The code is available at https://github.com/GeoX-Lab/Hook.
comment: 20 pages, 8 figures, 6 tables
☆ Physics-Informed Graph Neural Networks for Water Distribution Systems AAAI
Water distribution systems (WDS) are an integral part of critical infrastructure which is pivotal to urban development. As 70% of the world's population will likely live in urban environments in 2050, efficient simulation and planning tools for WDS play a crucial role in reaching UN's sustainable developmental goal (SDG) 6 - "Clean water and sanitation for all". In this realm, we propose a novel and efficient machine learning emulator, more precisely, a physics-informed deep learning (DL) model, for hydraulic state estimation in WDS. Using a recursive approach, our model only needs a few graph convolutional neural network (GCN) layers and employs an innovative algorithm based on message passing. Unlike conventional machine learning tasks, the model uses hydraulic principles to infer two additional hydraulic state features in the process of reconstructing the available ground truth feature in an unsupervised manner. To the best of our knowledge, this is the first DL approach to emulate the popular hydraulic simulator EPANET, utilizing no additional information. Like most DL models and unlike the hydraulic simulator, our model demonstrates vastly faster emulation times that do not increase drastically with the size of the WDS. Moreover, we achieve high accuracy on the ground truth and very similar results compared to the hydraulic simulator as demonstrated through experiments on five real-world WDS datasets.
comment: Extended version of the paper with the same title published at Proceedings of the AAAI Conference on Artificial Intelligence 2024
☆ PDNNet: PDN-Aware GNN-CNN Heterogeneous Network for Dynamic IR Drop Prediction
IR drop on the power delivery network (PDN) is closely related to PDN's configuration and cell current consumption. As the integrated circuit (IC) design is growing larger, dynamic IR drop simulation becomes computationally unaffordable and machine learning based IR drop prediction has been explored as a promising solution. Although CNN-based methods have been adapted to IR drop prediction task in several works, the shortcomings of overlooking PDN configuration is non-negligible. In this paper, we consider not only how to properly represent cell-PDN relation, but also how to model IR drop following its physical nature in the feature aggregation procedure. Thus, we propose a novel graph structure, PDNGraph, to unify the representations of the PDN structure and the fine-grained cell-PDN relation. We further propose a dual-branch heterogeneous network, PDNNet, incorporating two parallel GNN-CNN branches to favorably capture the above features during the learning process. Several key designs are presented to make the dynamic IR drop prediction highly effective and interpretable. We are the first work to apply graph structure to deep-learning based dynamic IR drop prediction method. Experiments show that PDNNet outperforms the state-of-the-art CNN-based methods by up to 39.3% reduction in prediction error and achieves 545x speedup compared to the commercial tool, which demonstrates the superiority of our method.
☆ Neural Architecture Search for Sentence Classification with BERT
Pre training of language models on large text corpora is common practice in Natural Language Processing. Following, fine tuning of these models is performed to achieve the best results on a variety of tasks. In this paper we question the common practice of only adding a single output layer as a classification head on top of the network. We perform an AutoML search to find architectures that outperform the current single layer at only a small compute cost. We validate our classification architecture on a variety of NLP benchmarks from the GLUE dataset.
☆ Efficient Heatmap-Guided 6-Dof Grasp Detection in Cluttered Scenes
Fast and robust object grasping in clutter is a crucial component of robotics. Most current works resort to the whole observed point cloud for 6-Dof grasp generation, ignoring the guidance information excavated from global semantics, thus limiting high-quality grasp generation and real-time performance. In this work, we show that the widely used heatmaps are underestimated in the efficiency of 6-Dof grasp generation. Therefore, we propose an effective local grasp generator combined with grasp heatmaps as guidance, which infers in a global-to-local semantic-to-point way. Specifically, Gaussian encoding and the grid-based strategy are applied to predict grasp heatmaps as guidance to aggregate local points into graspable regions and provide global semantic information. Further, a novel non-uniform anchor sampling mechanism is designed to improve grasp accuracy and diversity. Benefiting from the high-efficiency encoding in the image space and focusing on points in local graspable regions, our framework can perform high-quality grasp detection in real-time and achieve state-of-the-art results. In addition, real robot experiments demonstrate the effectiveness of our method with a success rate of 94% and a clutter completion rate of 100%. Our code is available at https://github.com/THU-VCLab/HGGD.
comment: Extensive results on GraspNet-1B dataset
☆ A Path Towards Legal Autonomy: An interoperable and explainable approach to extracting, transforming, loading and computing legal information using large language models, expert systems and Bayesian networks
Legal autonomy - the lawful activity of artificial intelligence agents - can be achieved in one of two ways. It can be achieved either by imposing constraints on AI actors such as developers, deployers and users, and on AI resources such as data, or by imposing constraints on the range and scope of the impact that AI agents can have on the environment. The latter approach involves encoding extant rules concerning AI driven devices into the software of AI agents controlling those devices (e.g., encoding rules about limitations on zones of operations into the agent software of an autonomous drone device). This is a challenge since the effectivity of such an approach requires a method of extracting, loading, transforming and computing legal information that would be both explainable and legally interoperable, and that would enable AI agents to reason about the law. In this paper, we sketch a proof of principle for such a method using large language models (LLMs), expert legal systems known as legal decision paths, and Bayesian networks. We then show how the proposed method could be applied to extant regulation in matters of autonomous cars, such as the California Vehicle Code.
☆ A Novel Behavior-Based Recommendation System for E-commerce
The majority of existing recommender systems rely on user ratings, which are limited by the lack of user collaboration and the sparsity problem. To address these issues, this study proposes a behavior-based recommender system that leverages customers' natural behaviors, such as browsing and clicking, on e-commerce platforms. The proposed recommendation system involves clustering active customers, determining neighborhoods, collecting similar users, calculating product reputation based on similar users, and recommending high-reputation products. To overcome the complexity of customer behaviors and traditional clustering methods, an unsupervised clustering approach based on product categories is developed to enhance the recommendation methodology. This study makes notable contributions in several aspects. Firstly, a groundbreaking behavior-based recommendation methodology is developed, incorporating customer behavior to generate accurate and tailored recommendations leading to improved customer satisfaction and engagement. Secondly, an original unsupervised clustering method, focusing on product categories, enables more precise clustering and facilitates accurate recommendations. Finally, an approach to determine neighborhoods for active customers within clusters is established, ensuring grouping of customers with similar behavioral patterns to enhance recommendation accuracy and relevance. The proposed recommendation methodology and clustering method contribute to improved recommendation performance, offering valuable insights for researchers and practitioners in the field of e-commerce recommendation systems. Additionally, the proposed method outperforms benchmark methods in experiments conducted using a behavior dataset from the well-known e-commerce site Alibaba.
☆ Improving Line Search Methods for Large Scale Neural Network Training
In recent studies, line search methods have shown significant improvements in the performance of traditional stochastic gradient descent techniques, eliminating the need for a specific learning rate schedule. In this paper, we identify existing issues in state-of-the-art line search methods, propose enhancements, and rigorously evaluate their effectiveness. We test these methods on larger datasets and more complex data domains than before. Specifically, we improve the Armijo line search by integrating the momentum term from ADAM in its search direction, enabling efficient large-scale training, a task that was previously prone to failure using Armijo line search methods. Our optimization approach outperforms both the previous Armijo implementation and tuned learning rate schedules for Adam. Our evaluation focuses on Transformers and CNNs in the domains of NLP and image data. Our work is publicly available as a Python package, which provides a hyperparameter free Pytorch optimizer.
☆ Faster Convergence for Transformer Fine-tuning with Line Search Methods
Recent works have shown that line search methods greatly increase performance of traditional stochastic gradient descent methods on a variety of datasets and architectures [1], [2]. In this work we succeed in extending line search methods to the novel and highly popular Transformer architecture and dataset domains in natural language processing. More specifically, we combine the Armijo line search with the Adam optimizer and extend it by subdividing the networks architecture into sensible units and perform the line search separately on these local units. Our optimization method outperforms the traditional Adam optimizer and achieves significant performance improvements for small data sets or small training budgets, while performing equal or better for other tested cases. Our work is publicly available as a python package, which provides a hyperparameter-free pytorch optimizer that is compatible with arbitrary network architectures.
☆ Impact of Employing Weather Forecast Data as Input to the Estimation of Evapotranspiration by Deep Neural Network Models
Reference Evapotranspiration (ET0) is a key parameter for designing smart irrigation scheduling, since it is related by a coefficient to the water needs of a crop. The United Nations Food and Agriculture Organization, proposed a standard method for ET0 computation (FAO56PM), based on the parameterization of the Penman-Monteith equation, that is widely adopted in the literature. To compute ET0 using the FAO56-PM method, four main weather parameters are needed: temperature, humidity, wind, and solar radiation (SR). One way to make daily ET0 estimations for future days is to use freely available weather forecast services (WFSs), where many meteorological parameters are estimated up to the next 15 days. A problem with this method is that currently, SR is not provided as a free forecast parameter on most of those online services or, normally, such forecasts present a financial cost penalty. For this reason, several ET0 estimation models using machine and deep learning were developed and presented in the literature, that use as input features a reduced set of carefully selected weather parameters, that are compatible with common freely available WFSs. However, most studies on this topic have only evaluated model performance using data from weather stations (WSs), without considering the effect of using weather forecast data. In this study, the performance of authors' previous models is evaluated when using weather forecast data from two online WFSs, in the following scenarios: (i) direct ET0 estimation by an ANN model, and (ii) estimate SR by ANN model, and then use that estimation for ET0 computation, using the FAO56-PM method. Employing data collected from two WFSs and a WS located in Vale do Lobo, Portugal, the latter approach achieved the best result, with a coefficient of determination (R2) ranging between 0.893 and 0.667, when considering forecasts up to 15 days.
comment: A partial version of the work submitted to ESRE/INTERNATIONAL CONFERENCE ON ENVIRONMENTAL SCIENCES AND RENEWABLE ENERGY
☆ Synthesizing EEG Signals from Event-Related Potential Paradigms with Conditional Diffusion Models
Data scarcity in the brain-computer interface field can be alleviated through the use of generative models, specifically diffusion models. While diffusion models have previously been successfully applied to electroencephalogram (EEG) data, existing models lack flexibility w.r.t.~sampling or require alternative representations of the EEG data. To overcome these limitations, we introduce a novel approach to conditional diffusion models that utilizes classifier-free guidance to directly generate subject-, session-, and class-specific EEG data. In addition to commonly used metrics, domain-specific metrics are employed to evaluate the specificity of the generated samples. The results indicate that the proposed model can generate EEG data that resembles real data for each subject, session, and class.
comment: submitted to 9th Graz BCI conference, 6 pages, 3 figures, first figure is split into two subfigures, 1 table
☆ Density-guided Translator Boosts Synthetic-to-Real Unsupervised Domain Adaptive Segmentation of 3D Point Clouds CVPR2024
3D synthetic-to-real unsupervised domain adaptive segmentation is crucial to annotating new domains. Self-training is a competitive approach for this task, but its performance is limited by different sensor sampling patterns (i.e., variations in point density) and incomplete training strategies. In this work, we propose a density-guided translator (DGT), which translates point density between domains, and integrates it into a two-stage self-training pipeline named DGT-ST. First, in contrast to existing works that simultaneously conduct data generation and feature/output alignment within unstable adversarial training, we employ the non-learnable DGT to bridge the domain gap at the input level. Second, to provide a well-initialized model for self-training, we propose a category-level adversarial network in stage one that utilizes the prototype to prevent negative transfer. Finally, by leveraging the designs above, a domain-mixed self-training method with source-aware consistency loss is proposed in stage two to narrow the domain gap further. Experiments on two synthetic-to-real segmentation tasks (SynLiDAR $\rightarrow$ semanticKITTI and SynLiDAR $\rightarrow$ semanticPOSS) demonstrate that DGT-ST outperforms state-of-the-art methods, achieving 9.4$\%$ and 4.3$\%$ mIoU improvements, respectively. Code is available at \url{https://github.com/yuan-zm/DGT-ST}.
comment: CVPR2024
☆ CoBOS: Constraint-Based Online Scheduler for Human-Robot Collaboration
Assembly processes involving humans and robots are challenging scenarios because the individual activities and access to shared workspace have to be coordinated. Fixed robot programs leave no room to diverge from a fixed protocol. Working on such a process can be stressful for the user and lead to ineffective behavior or failure. We propose a novel approach of online constraint-based scheduling in a reactive execution control framework facilitating behavior trees called CoBOS. This allows the robot to adapt to uncertain events such as delayed activity completions and activity selection (by the human). The user will experience less stress as the robotic coworkers adapt their behavior to best complement the human-selected activities to complete the common task. In addition to the improved working conditions, our algorithm leads to increased efficiency, even in highly uncertain scenarios. We evaluate our algorithm using a probabilistic simulation study with 56000 experiments. We outperform all baselines by a margin of 4-10%. Initial real robot experiments using a Franka Emika Panda robot and human tracking based on HTC Vive VR gloves look promising.
comment: 7 pages, 8 figures
☆ CoRAST: Towards Foundation Model-Powered Correlated Data Analysis in Resource-Constrained CPS and IoT
Foundation models (FMs) emerge as a promising solution to harness distributed and diverse environmental data by leveraging prior knowledge to understand the complicated temporal and spatial correlations within heterogeneous datasets. Unlike distributed learning frameworks such as federated learning, which often struggle with multimodal data, FMs can transform diverse inputs into embeddings. This process facilitates the integration of information from various modalities and the application of prior learning to new domains. However, deploying FMs in resource-constrained edge systems poses significant challenges. To this end, we introduce CoRAST, a novel learning framework that utilizes FMs for enhanced analysis of distributed, correlated heterogeneous data. Utilizing a server-based FM, CoRAST can exploit existing environment information to extract temporal, spatial, and cross-modal correlations among sensor data. This enables CoRAST to offer context-aware insights for localized client tasks through FM-powered global representation learning. Our evaluation on real-world weather dataset demonstrates CoRAST's ability to exploit correlated heterogeneous data through environmental representation learning to reduce the forecast errors by up to 50.3% compared to the baselines.
comment: accepted and to be published in 2024 IEEE International Workshop on Foundation Models for Cyber-Physical Systems & Internet of Things (FMSys)
☆ U-Sketch: An Efficient Approach for Sketch to Image Diffusion Models
Diffusion models have demonstrated remarkable performance in text-to-image synthesis, producing realistic and high resolution images that faithfully adhere to the corresponding text-prompts. Despite their great success, they still fall behind in sketch-to-image synthesis tasks, where in addition to text-prompts, the spatial layout of the generated images has to closely follow the outlines of certain reference sketches. Employing an MLP latent edge predictor to guide the spatial layout of the synthesized image by predicting edge maps at each denoising step has been recently proposed. Despite yielding promising results, the pixel-wise operation of the MLP does not take into account the spatial layout as a whole, and demands numerous denoising iterations to produce satisfactory images, leading to time inefficiency. To this end, we introduce U-Sketch, a framework featuring a U-Net type latent edge predictor, which is capable of efficiently capturing both local and global features, as well as spatial correlations between pixels. Moreover, we propose the addition of a sketch simplification network that offers the user the choice of preprocessing and simplifying input sketches for enhanced outputs. The experimental results, corroborated by user feedback, demonstrate that our proposed U-Net latent edge predictor leads to more realistic results, that are better aligned with the spatial outlines of the reference sketches, while drastically reducing the number of required denoising steps and, consequently, the overall execution time.
☆ BioMedLM: A 2.7B Parameter Language Model Trained On Biomedical Text
Models such as GPT-4 and Med-PaLM 2 have demonstrated impressive performance on a wide variety of biomedical NLP tasks. However, these models have hundreds of billions of parameters, are computationally expensive to run, require users to send their input data over the internet, and are trained on unknown data sources. Can smaller, more targeted models compete? To address this question, we build and release BioMedLM, a 2.7 billion parameter GPT-style autoregressive model trained exclusively on PubMed abstracts and full articles. When fine-tuned, BioMedLM can produce strong multiple-choice biomedical question-answering results competitive with much larger models, such as achieving a score of 57.3% on MedMCQA (dev) and 69.0% on the MMLU Medical Genetics exam. BioMedLM can also be fine-tuned to produce useful answers to patient questions on medical topics. This demonstrates that smaller models can potentially serve as transparent, privacy-preserving, economical and environmentally friendly foundations for particular NLP applications, such as in biomedicine. The model is available on the Hugging Face Hub: https://huggingface.co/stanford-crfm/BioMedLM.
comment: 23 pages
☆ A Channel-ensemble Approach: Unbiased and Low-variance Pseudo-labels is Critical for Semi-supervised Classification
Semi-supervised learning (SSL) is a practical challenge in computer vision. Pseudo-label (PL) methods, e.g., FixMatch and FreeMatch, obtain the State Of The Art (SOTA) performances in SSL. These approaches employ a threshold-to-pseudo-label (T2L) process to generate PLs by truncating the confidence scores of unlabeled data predicted by the self-training method. However, self-trained models typically yield biased and high-variance predictions, especially in the scenarios when a little labeled data are supplied. To address this issue, we propose a lightweight channel-based ensemble method to effectively consolidate multiple inferior PLs into the theoretically guaranteed unbiased and low-variance one. Importantly, our approach can be readily extended to any SSL framework, such as FixMatch or FreeMatch. Experimental results demonstrate that our method significantly outperforms state-of-the-art techniques on CIFAR10/100 in terms of effectiveness and efficiency.
☆ An Image Grid Can Be Worth a Video: Zero-shot Video Question Answering Using a VLM
Stimulated by the sophisticated reasoning capabilities of recent Large Language Models (LLMs), a variety of strategies for bridging video modality have been devised. A prominent strategy involves Video Language Models (VideoLMs), which train a learnable interface with video data to connect advanced vision encoders with LLMs. Recently, an alternative strategy has surfaced, employing readily available foundation models, such as VideoLMs and LLMs, across multiple stages for modality bridging. In this study, we introduce a simple yet novel strategy where only a single Vision Language Model (VLM) is utilized. Our starting point is the plain insight that a video comprises a series of images, or frames, interwoven with temporal information. The essence of video comprehension lies in adeptly managing the temporal aspects along with the spatial details of each frame. Initially, we transform a video into a single composite image by arranging multiple frames in a grid layout. The resulting single image is termed as an image grid. This format, while maintaining the appearance of a solitary image, effectively retains temporal information within the grid structure. Therefore, the image grid approach enables direct application of a single high-performance VLM without necessitating any video-data training. Our extensive experimental analysis across ten zero-shot video question answering benchmarks, including five open-ended and five multiple-choice benchmarks, reveals that the proposed Image Grid Vision Language Model (IG-VLM) surpasses the existing methods in nine out of ten benchmarks.
comment: Our code is available at https://github.com/imagegridworth/IG-VLM
☆ Leveraging Large Language Models for Relevance Judgments in Legal Case Retrieval
Collecting relevant judgments for legal case retrieval is a challenging and time-consuming task. Accurately judging the relevance between two legal cases requires a considerable effort to read the lengthy text and a high level of domain expertise to extract Legal Facts and make juridical judgments. With the advent of advanced large language models, some recent studies have suggested that it is promising to use LLMs for relevance judgment. Nonetheless, the method of employing a general large language model for reliable relevance judgments in legal case retrieval is yet to be thoroughly explored. To fill this research gap, we devise a novel few-shot workflow tailored to the relevant judgment of legal cases. The proposed workflow breaks down the annotation process into a series of stages, imitating the process employed by human annotators and enabling a flexible integration of expert reasoning to enhance the accuracy of relevance judgments. By comparing the relevance judgments of LLMs and human experts, we empirically show that we can obtain reliable relevance judgments with the proposed workflow. Furthermore, we demonstrate the capacity to augment existing legal case retrieval models through the synthesis of data generated by the large language model.
☆ Colour and Brush Stroke Pattern Recognition in Abstract Art using Modified Deep Convolutional Generative Adversarial Networks
Abstract Art is an immensely popular, discussed form of art that often has the ability to depict the emotions of an artist. Many researchers have made attempts to study abstract art in the form of edge detection, brush stroke and emotion recognition algorithms using machine and deep learning. This papers describes the study of a wide distribution of abstract paintings using Generative Adversarial Neural Networks(GAN). GANs have the ability to learn and reproduce a distribution enabling researchers and scientists to effectively explore and study the generated image space. However, the challenge lies in developing an efficient GAN architecture that overcomes common training pitfalls. This paper addresses this challenge by introducing a modified-DCGAN (mDCGAN) specifically designed for high-quality artwork generation. The approach involves a thorough exploration of the modifications made, delving into the intricate workings of DCGANs, optimisation techniques, and regularisation methods aimed at improving stability and realism in art generation enabling effective study of generated patterns. The proposed mDCGAN incorporates meticulous adjustments in layer configurations and architectural choices, offering tailored solutions to the unique demands of art generation while effectively combating issues like mode collapse and gradient vanishing. Further this paper explores the generated latent space by performing random walks to understand vector relationships between brush strokes and colours in the abstract art space and a statistical analysis of unstable outputs after a certain period of GAN training and compare its significant difference. These findings validate the effectiveness of the proposed approach, emphasising its potential to revolutionise the field of digital art generation and digital art ecosystem.
comment: 28 pages, 5 tables, 7 figures
☆ FTBC: Forward Temporal Bias Correction for Optimizing ANN-SNN Conversion
Spiking Neural Networks (SNNs) offer a promising avenue for energy-efficient computing compared with Artificial Neural Networks (ANNs), closely mirroring biological neural processes. However, this potential comes with inherent challenges in directly training SNNs through spatio-temporal backpropagation -- stemming from the temporal dynamics of spiking neurons and their discrete signal processing -- which necessitates alternative ways of training, most notably through ANN-SNN conversion. In this work, we introduce a lightweight Forward Temporal Bias Correction (FTBC) technique, aimed at enhancing conversion accuracy without the computational overhead. We ground our method on provided theoretical findings that through proper temporal bias calibration the expected error of ANN-SNN conversion can be reduced to be zero after each time step. We further propose a heuristic algorithm for finding the temporal bias only in the forward pass, thus eliminating the computational burden of backpropagation and we evaluate our method on CIFAR-10/100 and ImageNet datasets, achieving a notable increase in accuracy on all datasets. Codes are released at a GitHub repository.
☆ Generative Multi-modal Models are Good Class-Incremental Learners CVPR 2024
In class-incremental learning (CIL) scenarios, the phenomenon of catastrophic forgetting caused by the classifier's bias towards the current task has long posed a significant challenge. It is mainly caused by the characteristic of discriminative models. With the growing popularity of the generative multi-modal models, we would explore replacing discriminative models with generative ones for CIL. However, transitioning from discriminative to generative models requires addressing two key challenges. The primary challenge lies in transferring the generated textual information into the classification of distinct categories. Additionally, it requires formulating the task of CIL within a generative framework. To this end, we propose a novel generative multi-modal model (GMM) framework for class-incremental learning. Our approach directly generates labels for images using an adapted generative model. After obtaining the detailed text, we use a text encoder to extract text features and employ feature matching to determine the most similar label as the classification prediction. In the conventional CIL settings, we achieve significantly better results in long-sequence task scenarios. Under the Few-shot CIL setting, we have improved by at least 14\% accuracy over all the current state-of-the-art methods with significantly less forgetting. Our code is available at \url{https://github.com/DoubleClass/GMM}.
comment: Accepted at CVPR 2024
☆ Improving Attributed Text Generation of Large Language Models via Preference Learning
Large language models have been widely adopted in natural language processing, yet they face the challenge of generating unreliable content. Recent works aim to reduce misinformation and hallucinations by resorting to attribution as a means to provide evidence (i.e., citations). However, current attribution methods usually focus on the retrieval stage and automatic evaluation that neglect mirroring the citation mechanisms in human scholarly writing to bolster credibility. In this paper, we address these challenges by modelling the attribution task as preference learning and introducing an Automatic Preference Optimization (APO) framework. First, we create a curated collection for post-training with 6,330 examples by collecting and filtering from existing datasets. Second, considering the high cost of labelling preference data, we further propose an automatic method to synthesize attribution preference data resulting in 95,263 pairs. Moreover, inspired by the human citation process, we further propose a progressive preference optimization method by leveraging fine-grained information. Extensive experiments on three datasets (i.e., ASQA, StrategyQA, and ELI5) demonstrate that APO achieves state-of-the-art citation F1 with higher answer quality.
comment: 23 pages, 15 tables, 2 figures
☆ IIP-Mixer:Intra-Inter Patch Mixing Architecture for Battery Remaining Useful Life Prediction
Accurately estimating the Remaining Useful Life (RUL) of lithium-ion batteries is crucial for maintaining the safe and stable operation of rechargeable battery management systems. However, this task is often challenging due to the complex temporal dynamics involved. Recently, attention-based networks, such as Transformers and Informer, have been the popular architecture in time series forecasting. Despite their effectiveness, these models with abundant parameters necessitate substantial training time to unravel temporal patterns. To tackle these challenges, we propose a simple MLP-Mixer-based architecture named 'Intra-Inter Patch Mixer' (IIP-Mixer), which is an architecture based exclusively on multi-layer perceptrons (MLPs), extracting information by mixing operations along both intra-patch and inter-patch dimensions for battery RUL prediction. The proposed IIP-Mixer comprises parallel dual-head mixer layers: the intra-patch mixing MLP, capturing local temporal patterns in the short-term period, and the inter-patch mixing MLP, capturing global temporal patterns in the long-term period. Notably, to address the varying importance of features in RUL prediction, we introduce a weighted loss function in the MLP-Mixer-based architecture, marking the first time such an approach has been employed. Our experiments demonstrate that IIP-Mixer achieves competitive performance in battery RUL prediction, outperforming other popular time-series frameworks
☆ Intent-Aware DRL-Based Uplink Dynamic Scheduler for 5G-NR
We investigate the problem of supporting Industrial Internet of Things user equipment (IIoT UEs) with intent (i.e., requested quality of service (QoS)) and random traffic arrival. A deep reinforcement learning (DRL) based centralized dynamic scheduler for time-frequency resources is proposed to learn how to schedule the available communication resources among the IIoT UEs. The proposed scheduler leverages an RL framework to adapt to the dynamic changes in the wireless communication system and traffic arrivals. Moreover, a graph-based reduction scheme is proposed to reduce the state and action space of the RL framework to allow fast convergence and a better learning strategy. Simulation results demonstrate the effectiveness of the proposed intelligent scheduler in guaranteeing the expressed intent of IIoT UEs compared to several traditional scheduling schemes, such as round-robin, semi-static, and heuristic approaches. The proposed scheduler also outperforms the contention-free and contention-based schemes in maximizing the number of successfully computed tasks.
☆ Generating Diverse Agricultural Data for Vision-Based Farming Applications
We present a specialized procedural model for generating synthetic agricultural scenes, focusing on soybean crops, along with various weeds. This model is capable of simulating distinct growth stages of these plants, diverse soil conditions, and randomized field arrangements under varying lighting conditions. The integration of real-world textures and environmental factors into the procedural generation process enhances the photorealism and applicability of the synthetic data. Our dataset includes 12,000 images with semantic labels, offering a comprehensive resource for computer vision tasks in precision agriculture, such as semantic segmentation for autonomous weed control. We validate our model's effectiveness by comparing the synthetic data against real agricultural images, demonstrating its potential to significantly augment training data for machine learning models in agriculture. This approach not only provides a cost-effective solution for generating high-quality, diverse data but also addresses specific needs in agricultural vision tasks that are not fully covered by general-purpose models.
comment: 10 pages, 8 figures, 3 tables
☆ A Quantum Fuzzy-based Approach for Real-Time Detection of Solar Coronal Holes
The detection and analysis of the solar coronal holes (CHs) is an important field of study in the domain of solar physics. Mainly, it is required for the proper prediction of the geomagnetic storms which directly or indirectly affect various space and ground-based systems. For the detection of CHs till date, the solar scientist depends on manual hand-drawn approaches. However, with the advancement of image processing technologies, some automated image segmentation methods have been used for the detection of CHs. In-spite of this, fast and accurate detection of CHs are till a major issues. Here in this work, a novel quantum computing-based fast fuzzy c-mean technique has been developed for fast detection of the CHs region. The task has been carried out in two stages, in first stage the solar image has been segmented using a quantum computing based fast fuzzy c-mean (QCFFCM) and in the later stage the CHs has been extracted out from the segmented image based on image morphological operation. In the work, quantum computing has been used to optimize the cost function of the fast fuzzy c-mean (FFCM) algorithm, where quantum approximate optimization algorithm (QAOA) has been used to optimize the quadratic part of the cost function. The proposed method has been tested for 193 \AA{} SDO/AIA full-disk solar image datasets and has been compared with the existing techniques. The outcome shows the comparable performance of the proposed method with the existing one within a very lesser time.
comment: 14 pages, 5 figures, 3 tables
☆ LC-LLM: Explainable Lane-Change Intention and Trajectory Predictions with Large Language Models
To ensure safe driving in dynamic environments, autonomous vehicles should possess the capability to accurately predict the lane change intentions of surrounding vehicles in advance and forecast their future trajectories. Existing motion prediction approaches have ample room for improvement, particularly in terms of long-term prediction accuracy and interpretability. In this paper, we address these challenges by proposing LC-LLM, an explainable lane change prediction model that leverages the strong reasoning capabilities and self-explanation abilities of Large Language Models (LLMs). Essentially, we reformulate the lane change prediction task as a language modeling problem, processing heterogeneous driving scenario information in natural language as prompts for input into the LLM and employing a supervised fine-tuning technique to tailor the LLM specifically for our lane change prediction task. This allows us to utilize the LLM's powerful common sense reasoning abilities to understand complex interactive information, thereby improving the accuracy of long-term predictions. Furthermore, we incorporate explanatory requirements into the prompts in the inference stage. Therefore, our LC-LLM model not only can predict lane change intentions and trajectories but also provides explanations for its predictions, enhancing the interpretability. Extensive experiments on the large-scale highD dataset demonstrate the superior performance and interpretability of our LC-LLM in lane change prediction task. To the best of our knowledge, this is the first attempt to utilize LLMs for predicting lane change behavior. Our study shows that LLMs can encode comprehensive interaction information for driving behavior understanding.
☆ mALBERT: Is a Compact Multilingual BERT Model Still Worth It?
Within the current trend of Pretained Language Models (PLM), emerge more and more criticisms about the ethical andecological impact of such models. In this article, considering these critical remarks, we propose to focus on smallermodels, such as compact models like ALBERT, which are more ecologically virtuous than these PLM. However,PLMs enable huge breakthroughs in Natural Language Processing tasks, such as Spoken and Natural LanguageUnderstanding, classification, Question--Answering tasks. PLMs also have the advantage of being multilingual, and,as far as we know, a multilingual version of compact ALBERT models does not exist. Considering these facts, wepropose the free release of the first version of a multilingual compact ALBERT model, pre-trained using Wikipediadata, which complies with the ethical aspect of such a language model. We also evaluate the model against classicalmultilingual PLMs in classical NLP tasks. Finally, this paper proposes a rare study on the subword tokenizationimpact on language performances.
comment: The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation, May 2024, Torino, Italy
☆ Can LLMs Converse Formally? Automatically Assessing LLMs in Translating and Interpreting Formal Specifications
Stakeholders often describe system requirements using natural language which are then converted to formal syntax by a domain-expert leading to increased design costs. This paper assesses the capabilities of Large Language Models (LLMs) in converting between natural language descriptions and formal specifications. Existing work has evaluated the capabilities of LLMs in generating formal syntax such as source code but such experiments are typically hand-crafted and use problems that are likely to be in the training set of LLMs, and often require human-annotated datasets. We propose an approach that can use two copies of an LLM in conjunction with an off-the-shelf verifier to automatically evaluate its translation abilities without any additional human input. Our approach generates formal syntax using language grammars to automatically generate a dataset. We conduct an empirical evaluation to measure the accuracy of this translation task and show that SOTA LLMs cannot adequately solve this task, limiting their current utility in the design of complex systems.
☆ Chinese Offensive Language Detection:Current Status and Future Directions
Despite the considerable efforts being made to monitor and regulate user-generated content on social media platforms, the pervasiveness of offensive language, such as hate speech or cyberbullying, in the digital space remains a significant challenge. Given the importance of maintaining a civilized and respectful online environment, there is an urgent and growing need for automatic systems capable of detecting offensive speech in real time. However, developing effective systems for processing languages such as Chinese presents a significant challenge, owing to the language's complex and nuanced nature, which makes it difficult to process automatically. This paper provides a comprehensive overview of offensive language detection in Chinese, examining current benchmarks and approaches and highlighting specific models and tools for addressing the unique challenges of detecting offensive language in this complex language. The primary objective of this survey is to explore the existing techniques and identify potential avenues for further research that can address the cultural and linguistic complexities of Chinese.
☆ A thermodynamically consistent physics-informed deep learning material model for short fiber/polymer nanocomposites
This work proposes a physics-informed deep learning (PIDL)-based constitutive model for investigating the viscoelastic-viscoplastic behavior of short fiber-reinforced nanoparticle-filled epoxies under various ambient conditions. The deep-learning model is trained to enforce thermodynamic principles, leading to a thermodynamically consistent constitutive model. To accomplish this, a long short-term memory network is combined with a feed-forward neural network to predict internal variables required for characterizing the internal dissipation of the nanocomposite materials. In addition, another feed-forward neural network is used to indicate the free-energy function, which enables defining the thermodynamic state of the entire system. The PIDL model is initially developed for the three-dimensional case by generating synthetic data from a classical constitutive model. The model is then trained by extracting the data directly from cyclic loading-unloading experimental tests. Numerical examples show that the PIDL model can accurately predict the mechanical behavior of epoxy-based nanocomposites for different volume fractions of fibers and nanoparticles under various hygrothermal conditions.
comment: arXiv admin note: text overlap with arXiv:2305.08102
☆ A Recommender System for NFT Collectibles with Item Feature AAAI 2023
Recommender systems have been actively studied and applied in various domains to deal with information overload. Although there are numerous studies on recommender systems for movies, music, and e-commerce, comparatively less attention has been paid to the recommender system for NFTs despite the continuous growth of the NFT market. This paper presents a recommender system for NFTs that utilizes a variety of data sources, from NFT transaction records to external item features, to generate precise recommendations that cater to individual preferences. We develop a data-efficient graph-based recommender system to efficiently capture the complex relationship between each item and users and generate node(item) embeddings which incorporate both node feature information and graph structure. Furthermore, we exploit inputs beyond user-item interactions, such as image feature, text feature, and price feature. Numerical experiments verify the performance of the graph-based recommender system improves significantly after utilizing all types of item features as side information, thereby outperforming all other baselines.
comment: Presented at the AAAI 2023 Bridge on AI for Financial Services (https://sites.google.com/view/aaai-ai-fin/home)
☆ Selective Mixup Fine-Tuning for Optimizing Non-Decomposable Objectives ICLR 2024
The rise in internet usage has led to the generation of massive amounts of data, resulting in the adoption of various supervised and semi-supervised machine learning algorithms, which can effectively utilize the colossal amount of data to train models. However, before deploying these models in the real world, these must be strictly evaluated on performance measures like worst-case recall and satisfy constraints such as fairness. We find that current state-of-the-art empirical techniques offer sub-optimal performance on these practical, non-decomposable performance objectives. On the other hand, the theoretical techniques necessitate training a new model from scratch for each performance objective. To bridge the gap, we propose SelMix, a selective mixup-based inexpensive fine-tuning technique for pre-trained models, to optimize for the desired objective. The core idea of our framework is to determine a sampling distribution to perform a mixup of features between samples from particular classes such that it optimizes the given objective. We comprehensively evaluate our technique against the existing empirical and theoretically principled methods on standard benchmark datasets for imbalanced classification. We find that proposed SelMix fine-tuning significantly improves the performance for various practical non-decomposable objectives across benchmarks.
comment: ICLR 2024 SpotLight
☆ GeNet: A Graph Neural Network-based Anti-noise Task-Oriented Semantic Communication Paradigm
Traditional approaches to semantic communication tasks rely on the knowledge of the signal-to-noise ratio (SNR) to mitigate channel noise. However, these methods necessitate training under specific SNR conditions, entailing considerable time and computational resources. In this paper, we propose GeNet, a Graph Neural Network (GNN)-based paradigm for semantic communication aimed at combating noise, thereby facilitating Task-Oriented Communication (TOC). We propose a novel approach where we first transform the input data image into graph structures. Then we leverage a GNN-based encoder to extract semantic information from the source data. This extracted semantic information is then transmitted through the channel. At the receiver's end, a GNN-based decoder is utilized to reconstruct the relevant semantic information from the source data for TOC. Through experimental evaluation, we show GeNet's effectiveness in anti-noise TOC while decoupling the SNR dependency. We further evaluate GeNet's performance by varying the number of nodes, revealing its versatility as a new paradigm for semantic communication. Additionally, we show GeNet's robustness to geometric transformations by testing it with different rotation angles, without resorting to data augmentation.
☆ Few-Shot Recalibration of Language Models
Recent work has uncovered promising ways to extract well-calibrated confidence estimates from language models (LMs), where the model's confidence score reflects how likely it is to be correct. However, while LMs may appear well-calibrated over broad distributions, this often hides significant miscalibration within narrower slices (e.g., systemic over-confidence in math can balance out systemic under-confidence in history, yielding perfect calibration in aggregate). To attain well-calibrated confidence estimates for any slice of a distribution, we propose a new framework for few-shot slice-specific recalibration. Specifically, we train a recalibration model that takes in a few unlabeled examples from any given slice and predicts a curve that remaps confidence scores to be more accurate for that slice. Our trained model can recalibrate for arbitrary new slices, without using any labeled data from that slice. This enables us to identify domain-specific confidence thresholds above which the LM's predictions can be trusted, and below which it should abstain. Experiments show that our few-shot recalibrator consistently outperforms existing calibration methods, for instance improving calibration error for PaLM2-Large on MMLU by 16%, as compared to temperature scaling.
comment: preprint
☆ Identification and Uses of Deep Learning Backbones via Pattern Mining SDM24
Deep learning is extensively used in many areas of data mining as a black-box method with impressive results. However, understanding the core mechanism of how deep learning makes predictions is a relatively understudied problem. Here we explore the notion of identifying a backbone of deep learning for a given group of instances. A group here can be instances of the same class or even misclassified instances of the same class. We view each instance for a given group as activating a subset of neurons and attempt to find a subgraph of neurons associated with a given concept/group. We formulate this problem as a set cover style problem and show it is intractable and presents a highly constrained integer linear programming (ILP) formulation. As an alternative, we explore a coverage-based heuristic approach related to pattern mining, and show it converges to a Pareto equilibrium point of the ILP formulation. Experimentally we explore these backbones to identify mistakes and improve performance, explanation, and visualization. We demonstrate application-based results using several challenging data sets, including Bird Audio Detection (BAD) Challenge and Labeled Faces in the Wild (LFW), as well as the classic MNIST data.
comment: 9 pages, 6 figures, published SIAM SDM24
☆ DSF-GAN: DownStream Feedback Generative Adversarial Network
Utility and privacy are two crucial measurements of the quality of synthetic tabular data. While significant advancements have been made in privacy measures, generating synthetic samples with high utility remains challenging. To enhance the utility of synthetic samples, we propose a novel architecture called the DownStream Feedback Generative Adversarial Network (DSF-GAN). This approach incorporates feedback from a downstream prediction model during training to augment the generator's loss function with valuable information. Thus, DSF-GAN utilizes a downstream prediction task to enhance the utility of synthetic samples. To evaluate our method, we tested it using two popular datasets. Our experiments demonstrate improved model performance when training on synthetic samples generated by DSF-GAN, compared to those generated by the same GAN architecture without feedback. The evaluation was conducted on the same validation set comprising real samples. All code and datasets used in this research will be made openly available for ease of reproduction.
☆ Enhancing Generative Class Incremental Learning Performance with Model Forgetting Approach
This study presents a novel approach to Generative Class Incremental Learning (GCIL) by introducing the forgetting mechanism, aimed at dynamically managing class information for better adaptation to streaming data. GCIL is one of the hot topics in the field of computer vision, and this is considered one of the crucial tasks in society, specifically the continual learning of generative models. The ability to forget is a crucial brain function that facilitates continual learning by selectively discarding less relevant information for humans. However, in the field of machine learning models, the concept of intentionally forgetting has not been extensively investigated. In this study we aim to bridge this gap by incorporating the forgetting mechanisms into GCIL, thereby examining their impact on the models' ability to learn in continual learning. Through our experiments, we have found that integrating the forgetting mechanisms significantly enhances the models' performance in acquiring new knowledge, underscoring the positive role that strategic forgetting plays in the process of continual learning.
☆ Manipulating Neural Path Planners via Slight Perturbations
Data-driven neural path planners are attracting increasing interest in the robotics community. However, their neural network components typically come as black boxes, obscuring their underlying decision-making processes. Their black-box nature exposes them to the risk of being compromised via the insertion of hidden malicious behaviors. For example, an attacker may hide behaviors that, when triggered, hijack a delivery robot by guiding it to a specific (albeit wrong) destination, trapping it in a predefined region, or inducing unnecessary energy expenditure by causing the robot to repeatedly circle a region. In this paper, we propose a novel approach to specify and inject a range of hidden malicious behaviors, known as backdoors, into neural path planners. Our approach provides a concise but flexible way to define these behaviors, and we show that hidden behaviors can be triggered by slight perturbations (e.g., inserting a tiny unnoticeable object), that can nonetheless significantly compromise their integrity. We also discuss potential techniques to identify these backdoors aimed at alleviating such risks. We demonstrate our approach on both sampling-based and search-based neural path planners.
☆ Beyond Embeddings: The Promise of Visual Table in Multi-Modal Models
Visual representation learning has been a cornerstone in computer vision, evolving from supervised learning with human-annotated labels to aligning image-text pairs from the Internet. Despite recent advancements in multi-modal large language models (MLLMs), the visual representations they rely on, such as CLIP embeddings, often lack access to external world knowledge critical for real-world visual reasoning. In this work, we propose Visual Table, a novel visual representation tailored for MLLMs. It provides hierarchical text descriptions of holistic visual scenes, consisting of a scene description and multiple object-centric descriptions that encompass categories, attributes, and knowledge at instance level. We further develop a scalable generator for visual table generation and train it on small-scale annotations from GPT4V. Extensive evaluations demonstrate that, with generated visual tables as additional visual representations, our model can consistently outperform the state-of-the-art (SOTA) MLLMs across diverse benchmarks. When visual tables serve as standalone visual representations, our model can closely match or even beat the SOTA MLLMs that are built on CLIP visual embeddings. Our code is available at https://github.com/LaVi-Lab/Visual-Table.
comment: Project page: https://github.com/LaVi-Lab/Visual-Table
☆ Boosting Conversational Question Answering with Fine-Grained Retrieval-Augmentation and Self-Check
Retrieval-Augmented Generation (RAG) aims to generate more reliable and accurate responses, by augmenting large language models (LLMs) with the external vast and dynamic knowledge. Most previous work focuses on using RAG for single-round question answering, while how to adapt RAG to the complex conversational setting wherein the question is interdependent on the preceding context is not well studied. In this paper, we propose a conversation-level RAG approach, which incorporates fine-grained retrieval augmentation and self-check for conversational question answering (CQA). In particular, our approach consists of three components, namely conversational question refiner, fine-grained retriever and self-check based response generator, which work collaboratively for question understanding and relevant information acquisition in conversational settings. Extensive experiments demonstrate the great advantages of our approach over the state-of-the-art baselines. Moreover, we also release a Chinese CQA dataset with new features including reformulated question, extracted keyword, retrieved paragraphs and their helpfulness, which facilitates further researches in RAG enhanced CQA.
☆ NeuSDFusion: A Spatial-Aware Generative Model for 3D Shape Completion, Reconstruction, and Generation
3D shape generation aims to produce innovative 3D content adhering to specific conditions and constraints. Existing methods often decompose 3D shapes into a sequence of localized components, treating each element in isolation without considering spatial consistency. As a result, these approaches exhibit limited versatility in 3D data representation and shape generation, hindering their ability to generate highly diverse 3D shapes that comply with the specified constraints. In this paper, we introduce a novel spatial-aware 3D shape generation framework that leverages 2D plane representations for enhanced 3D shape modeling. To ensure spatial coherence and reduce memory usage, we incorporate a hybrid shape representation technique that directly learns a continuous signed distance field representation of the 3D shape using orthogonal 2D planes. Additionally, we meticulously enforce spatial correspondences across distinct planes using a transformer-based autoencoder structure, promoting the preservation of spatial relationships in the generated 3D shapes. This yields an algorithm that consistently outperforms state-of-the-art 3D shape generation methods on various tasks, including unconditional shape generation, multi-modal shape completion, single-view reconstruction, and text-to-shape synthesis.
☆ Large Language Models Need Consultants for Reasoning: Becoming an Expert in a Complex Human System Through Behavior Simulation
Large language models (LLMs), in conjunction with various reasoning reinforcement methodologies, have demonstrated remarkable capabilities comparable to humans in fields such as mathematics, law, coding, common sense, and world knowledge. In this paper, we delve into the reasoning abilities of LLMs within complex human systems. We propose a novel reasoning framework, termed ``Mosaic Expert Observation Wall'' (MEOW) exploiting generative-agents-based simulation technique. In the MEOW framework, simulated data are utilized to train an expert model concentrating ``experience'' about a specific task in each independent time of simulation. It is the accumulated ``experience'' through the simulation that makes for an expert on a task in a complex human system. We conduct the experiments within a communication game that mirrors real-world security scenarios. The results indicate that our proposed methodology can cooperate with existing methodologies to enhance the reasoning abilities of LLMs in complex human systems.
☆ A Transformer-Based Framework for Payload Malware Detection and Classification
As malicious cyber threats become more sophisticated in breaching computer networks, the need for effective intrusion detection systems (IDSs) becomes crucial. Techniques such as Deep Packet Inspection (DPI) have been introduced to allow IDSs analyze the content of network packets, providing more context for identifying potential threats. IDSs traditionally rely on using anomaly-based and signature-based detection techniques to detect unrecognized and suspicious activity. Deep learning techniques have shown great potential in DPI for IDSs due to their efficiency in learning intricate patterns from the packet content being transmitted through the network. In this paper, we propose a revolutionary DPI algorithm based on transformers adapted for the purpose of detecting malicious traffic with a classifier head. Transformers learn the complex content of sequence data and generalize them well to similar scenarios thanks to their self-attention mechanism. Our proposed method uses the raw payload bytes that represent the packet contents and is deployed as man-in-the-middle. The payload bytes are used to detect malicious packets and classify their types. Experimental results on the UNSW-NB15 and CIC-IOT23 datasets demonstrate that our transformer-based model is effective in distinguishing malicious from benign traffic in the test dataset, attaining an average accuracy of 79\% using binary classification and 72\% on the multi-classification experiment, both using solely payload bytes.
☆ From Two-Dimensional to Three-Dimensional Environment with Q-Learning: Modeling Autonomous Navigation with Reinforcement Learning and no Libraries
Reinforcement learning (RL) algorithms have become indispensable tools in artificial intelligence, empowering agents to acquire optimal decision-making policies through interactions with their environment and feedback mechanisms. This study explores the performance of RL agents in both two-dimensional (2D) and three-dimensional (3D) environments, aiming to research the dynamics of learning across different spatial dimensions. A key aspect of this investigation is the absence of pre-made libraries for learning, with the algorithm developed exclusively through computational mathematics. The methodological framework centers on RL principles, employing a Q-learning agent class and distinct environment classes tailored to each spatial dimension. The research aims to address the question: How do reinforcement learning agents adapt and perform in environments of varying spatial dimensions, particularly in 2D and 3D settings? Through empirical analysis, the study evaluates agents' learning trajectories and adaptation processes, revealing insights into the efficacy of RL algorithms in navigating complex, multi-dimensional spaces. Reflections on the findings prompt considerations for future research, particularly in understanding the dynamics of learning in higher-dimensional environments.
☆ Leveraging Large Language Models for Fuzzy String Matching in Political Science
Fuzzy string matching remains a key issue when political scientists combine data from different sources. Existing matching methods invariably rely on string distances, such as Levenshtein distance and cosine similarity. As such, they are inherently incapable of matching strings that refer to the same entity with different names such as ''JP Morgan'' and ''Chase Bank'', ''DPRK'' and ''North Korea'', ''Chuck Fleischmann (R)'' and ''Charles Fleischmann (R)''. In this letter, we propose to use large language models to entirely sidestep this problem in an easy and intuitive manner. Extensive experiments show that our proposed methods can improve the state of the art by as much as 39% in terms of average precision while being substantially easier and more intuitive to use by political scientists. Moreover, our results are robust against various temperatures. We further note that enhanced prompting can lead to additional performance improvements.
comment: 7 pages, 2 figures, 1 table;
☆ Preference-Based Planning in Stochastic Environments: From Partially-Ordered Temporal Goals to Most Preferred Policies
Human preferences are not always represented via complete linear orders: It is natural to employ partially-ordered preferences for expressing incomparable outcomes. In this work, we consider decision-making and probabilistic planning in stochastic systems modeled as Markov decision processes (MDPs), given a partially ordered preference over a set of temporally extended goals. Specifically, each temporally extended goal is expressed using a formula in Linear Temporal Logic on Finite Traces (LTL$_f$). To plan with the partially ordered preference, we introduce order theory to map a preference over temporal goals to a preference over policies for the MDP. Accordingly, a most preferred policy under a stochastic ordering induces a stochastic nondominated probability distribution over the finite paths in the MDP. To synthesize a most preferred policy, our technical approach includes two key steps. In the first step, we develop a procedure to transform a partially ordered preference over temporal goals into a computational model, called preference automaton, which is a semi-automaton with a partial order over acceptance conditions. In the second step, we prove that finding a most preferred policy is equivalent to computing a Pareto-optimal policy in a multi-objective MDP that is constructed from the original MDP, the preference automaton, and the chosen stochastic ordering relation. Throughout the paper, we employ running examples to illustrate the proposed preference specification and solution approaches. We demonstrate the efficacy of our algorithm using these examples, providing detailed analysis, and then discuss several potential future directions.
comment: arXiv admin note: substantial text overlap with arXiv:2209.12267
☆ Long and Short-Term Constraints Driven Safe Reinforcement Learning for Autonomous Driving
Reinforcement learning (RL) has been widely used in decision-making tasks, but it cannot guarantee the agent's safety in the training process due to the requirements of interaction with the environment, which seriously limits its industrial applications such as autonomous driving. Safe RL methods are developed to handle this issue by constraining the expected safety violation costs as a training objective, but they still permit unsafe state occurrence, which is unacceptable in autonomous driving tasks. Moreover, these methods are difficult to achieve a balance between the cost and return expectations, which leads to learning performance degradation for the algorithms. In this paper, we propose a novel algorithm based on the long and short-term constraints (LSTC) for safe RL. The short-term constraint aims to guarantee the short-term state safety that the vehicle explores, while the long-term constraint ensures the overall safety of the vehicle throughout the decision-making process. In addition, we develop a safe RL method with dual-constraint optimization based on the Lagrange multiplier to optimize the training process for end-to-end autonomous driving. Comprehensive experiments were conducted on the MetaDrive simulator. Experimental results demonstrate that the proposed method achieves higher safety in continuous state and action tasks, and exhibits higher exploration performance in long-distance decision-making tasks compared with state-of-the-art methods.
☆ An Evolutionary Network Architecture Search Framework with Adaptive Multimodal Fusion for Hand Gesture Recognition
Hand gesture recognition (HGR) based on multimodal data has attracted considerable attention owing to its great potential in applications. Various manually designed multimodal deep networks have performed well in multimodal HGR (MHGR), but most of existing algorithms require a lot of expert experience and time-consuming manual trials. To address these issues, we propose an evolutionary network architecture search framework with the adaptive multimodel fusion (AMF-ENAS). Specifically, we design an encoding space that simultaneously considers fusion positions and ratios of the multimodal data, allowing for the automatic construction of multimodal networks with different architectures through decoding. Additionally, we consider three input streams corresponding to intra-modal surface electromyography (sEMG), intra-modal accelerometer (ACC), and inter-modal sEMG-ACC. To automatically adapt to various datasets, the ENAS framework is designed to automatically search a MHGR network with appropriate fusion positions and ratios. To the best of our knowledge, this is the first time that ENAS has been utilized in MHGR to tackle issues related to the fusion position and ratio of multimodal data. Experimental results demonstrate that AMF-ENAS achieves state-of-the-art performance on the Ninapro DB2, DB3, and DB7 datasets.
☆ Exploring the Privacy Protection Capabilities of Chinese Large Language Models
Large language models (LLMs), renowned for their impressive capabilities in various tasks, have significantly advanced artificial intelligence. Yet, these advancements have raised growing concerns about privacy and security implications. To address these issues and explain the risks inherent in these models, we have devised a three-tiered progressive framework tailored for evaluating privacy in language systems. This framework consists of progressively complex and in-depth privacy test tasks at each tier. Our primary objective is to comprehensively evaluate the sensitivity of large language models to private information, examining how effectively they discern, manage, and safeguard sensitive data in diverse scenarios. This systematic evaluation helps us understand the degree to which these models comply with privacy protection guidelines and the effectiveness of their inherent safeguards against privacy breaches. Our observations indicate that existing Chinese large language models universally show privacy protection shortcomings. It seems that at the moment this widespread issue is unavoidable and may pose corresponding privacy risks in applications based on these models.
comment: 11 pages
☆ EndToEndML: An Open-Source End-to-End Pipeline for Machine Learning Applications
Artificial intelligence (AI) techniques are widely applied in the life sciences. However, applying innovative AI techniques to understand and deconvolute biological complexity is hindered by the learning curve for life science scientists to understand and use computing languages. An open-source, user-friendly interface for AI models, that does not require programming skills to analyze complex biological data will be extremely valuable to the bioinformatics community. With easy access to different sequencing technologies and increased interest in different 'omics' studies, the number of biological datasets being generated has increased and analyzing these high-throughput datasets is computationally demanding. The majority of AI libraries today require advanced programming skills as well as machine learning, data preprocessing, and visualization skills. In this research, we propose a web-based end-to-end pipeline that is capable of preprocessing, training, evaluating, and visualizing machine learning (ML) models without manual intervention or coding expertise. By integrating traditional machine learning and deep neural network models with visualizations, our library assists in recognizing, classifying, clustering, and predicting a wide range of multi-modal, multi-sensor datasets, including images, languages, and one-dimensional numerical data, for drug discovery, pathogen classification, and medical diagnostics.
comment: 2024 7th International Conference on Information and Computer Technologies (ICICT)
☆ Looking Beyond What You See: An Empirical Analysis on Subgroup Intersectional Fairness for Multi-label Chest X-ray Classification Using Social Determinants of Racial Health Inequities ICCV
There has been significant progress in implementing deep learning models in disease diagnosis using chest X- rays. Despite these advancements, inherent biases in these models can lead to disparities in prediction accuracy across protected groups. In this study, we propose a framework to achieve accurate diagnostic outcomes and ensure fairness across intersectional groups in high-dimensional chest X- ray multi-label classification. Transcending traditional protected attributes, we consider complex interactions within social determinants, enabling a more granular benchmark and evaluation of fairness. We present a simple and robust method that involves retraining the last classification layer of pre-trained models using a balanced dataset across groups. Additionally, we account for fairness constraints and integrate class-balanced fine-tuning for multi-label settings. The evaluation of our method on the MIMIC-CXR dataset demonstrates that our framework achieves an optimal tradeoff between accuracy and fairness compared to baseline methods.
comment: ICCV CVAMD 2023
☆ SCANet: Correcting LEGO Assembly Errors with Self-Correct Assembly Network
Autonomous assembly in robotics and 3D vision presents significant challenges, particularly in ensuring assembly correctness. Presently, predominant methods such as MEPNet focus on assembling components based on manually provided images. However, these approaches often fall short in achieving satisfactory results for tasks requiring long-term planning. Concurrently, we observe that integrating a self-correction module can partially alleviate such issues. Motivated by this concern, we introduce the single-step assembly error correction task, which involves identifying and rectifying misassembled components. To support research in this area, we present the LEGO Error Correction Assembly Dataset (LEGO-ECA), comprising manual images for assembly steps and instances of assembly failures. Additionally, we propose the Self-Correct Assembly Network (SCANet), a novel method to address this task. SCANet treats assembled components as queries, determining their correctness in manual images and providing corrections when necessary. Finally, we utilize SCANet to correct the assembly results of MEPNet. Experimental results demonstrate that SCANet can identify and correct MEPNet's misassembled results, significantly improving the correctness of assembly. Our code and dataset are available at https://github.com/Yaser-wyx/SCANet.
Can AI Models Appreciate Document Aesthetics? An Exploration of Legibility and Layout Quality in Relation to Prediction Confidence
A well-designed document communicates not only through its words but also through its visual eloquence. Authors utilize aesthetic elements such as colors, fonts, graphics, and layouts to shape the perception of information. Thoughtful document design, informed by psychological insights, enhances both the visual appeal and the comprehension of the content. While state-of-the-art document AI models demonstrate the benefits of incorporating layout and image data, it remains unclear whether the nuances of document aesthetics are effectively captured. To bridge the gap between human cognition and AI interpretation of aesthetic elements, we formulated hypotheses concerning AI behavior in document understanding tasks, specifically anchored in document design principles. With a focus on legibility and layout quality, we tested four aspects of aesthetic effects: noise, font-size contrast, alignment, and complexity, on model confidence using correlational analysis. The results and observations highlight the value of model analysis rooted in document design theories. Our work serves as a trailhead for further studies and we advocate for continued research in this topic to deepen our understanding of how AI interprets document aesthetics.
☆ Mechanisms of non-factual hallucinations in language models
State-of-the-art language models (LMs) sometimes generate non-factual hallucinations that misalign with world knowledge. Despite extensive efforts to detect and mitigate hallucinations, understanding their internal mechanisms remains elusive. Our study investigates the mechanistic causes of hallucination, specifically non-factual ones where the LM incorrectly predicts object attributes in response to subject-relation queries. With causal mediation analysis and embedding space projection, we identify two general mechanistic causes of hallucinations shared across LMs of various scales and designs: 1) insufficient subject attribute knowledge in lower layer MLPs, and 2) failing to select the correct object attribute in upper layer attention heads and MLPs. These two mechanisms exhibit varying degrees of subject-object association, predictive uncertainty and perturbation robustness. Additionally, we scrutinize LM pre-training checkpoints, revealing distinct learning dynamics for the two mechanistic causes of hallucinations. We also highlight how attribution features from our causal analysis can effectively construct hallucination detectors. Our work proposes a mechanistic understanding of LM factual errors.
♻ ☆ A 4D Hybrid Algorithm to Scale Parallel Training to Thousands of GPUs
Large communication costs are a critical bottleneck in training state-of-the-art neural networks on distributed systems. This paper introduces AxoNN, a novel four-dimensional (4D) parallelization approach, inspired by Agarwal's algorithm for matrix multiplication, for parallelizing tensor computations in deep learning, AxoNN employs two key strategies to minimize communication overhead. First, we optimize communication by overlapping expensive collective operations (reduce-scatter, all-gather, all-reduce) with computations. Our experiments with a 20-billion parameter transformer model demonstrate that these optimizations deliver nearly 53\% improvement. Second, we present an analytical model to assist users in identifying communication-minimizing configurations within the vast search space defined by our 4D algorithm. This model empowers practitioners by simplifying the tuning process for their specific training workloads. When training an 80-billion parameter model on 1024 GPUs of Perlmutter, AxoNN surpasses Megatron-LM, a state-of-the-art framework, by a significant 26%. Additionally, it achieves 57% of the theoretical peak FLOP/s.
♻ ☆ Shifting to Machine Supervision: Annotation-Efficient Semi and Self-Supervised Learning for Automatic Medical Image Segmentation and Classification
Advancements in clinical treatment are increasingly constrained by the limitations of supervised learning techniques, which depend heavily on large volumes of annotated data. The annotation process is not only costly but also demands substantial time from clinical specialists. Addressing this issue, we introduce the S4MI (Self-Supervision and Semi-Supervision for Medical Imaging) pipeline, a novel approach that leverages advancements in self-supervised and semi-supervised learning. These techniques engage in auxiliary tasks that do not require labeling, thus simplifying the scaling of machine supervision compared to fully-supervised methods. Our study benchmarks these techniques on three distinct medical imaging datasets to evaluate their effectiveness in classification and segmentation tasks. Notably, we observed that self supervised learning significantly surpassed the performance of supervised methods in the classification of all evaluated datasets. Remarkably, the semi-supervised approach demonstrated superior outcomes in segmentation, outperforming fully-supervised methods while using 50% fewer labels across all datasets. In line with our commitment to contributing to the scientific community, we have made the S4MI code openly accessible, allowing for broader application and further development of these methods.
comment: Seventeen pages (incl. references), five figures, and one table. (Under Review)
♻ ☆ Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimization
Large Language Models exhibit robust problem-solving capabilities for diverse tasks. However, most LLM-based agents are designed as specific task solvers with sophisticated prompt engineering, rather than agents capable of learning and evolving through interactions. These task solvers necessitate manually crafted prompts to inform task rules and regulate LLM behaviors, inherently incapacitating to address complex dynamic scenarios e.g., large interactive games. In light of this, we propose Agent-Pro: an LLM-based Agent with Policy-level Reflection and Optimization that can learn a wealth of expertise from interactive experiences and progressively elevate its behavioral policy. Specifically, it involves a dynamic belief generation and reflection process for policy evolution. Rather than action-level reflection, Agent-Pro iteratively reflects on past trajectories and beliefs, fine-tuning its irrational beliefs for a better policy. Moreover, a depth-first search is employed for policy optimization, ensuring continual enhancement in policy payoffs. Agent-Pro is evaluated across two games: Blackjack and Texas Hold'em, outperforming vanilla LLM and specialized models. Our results show Agent-Pro can learn and evolve in complex and dynamic scenes, which also benefits numerous LLM-based applications.
comment: LLM-based Agent
♻ ☆ Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives
The reflection capacity of Large Language Model (LLM) has garnered extensive attention. A post-hoc prompting strategy, e.g., reflexion and self-refine, refines LLM's response based on self-evaluated or external feedback. However, recent research indicates without external feedback, LLM's intrinsic reflection is unstable. Our investigation unveils that the key bottleneck is the quality of the self-evaluated feedback. We find LLMs often exhibit overconfidence or high randomness when self-evaluate, offering stubborn or inconsistent feedback, which causes poor reflection. To remedy this, we advocate Self-Contrast: It adaptively explores diverse solving perspectives tailored to the request, contrasts the differences, and summarizes these discrepancies into a checklist which could be used to re-examine and eliminate discrepancies. Our method endows LLM with diverse perspectives to alleviate stubborn biases. Moreover, their discrepancies indicate potential errors or inherent uncertainties that LLM often overlooks. Reflecting upon these can catalyze more accurate and stable reflection. Experiments conducted on a series of reasoning and translation tasks with different LLMs serve to underscore the effectiveness and generality of our strategy.
♻ ☆ Generalization Bounds: Perspectives from Information Theory and PAC-Bayes
A fundamental question in theoretical machine learning is generalization. Over the past decades, the PAC-Bayesian approach has been established as a flexible framework to address the generalization capabilities of machine learning algorithms, and design new ones. Recently, it has garnered increased interest due to its potential applicability for a variety of learning algorithms, including deep neural networks. In parallel, an information-theoretic view of generalization has developed, wherein the relation between generalization and various information measures has been established. This framework is intimately connected to the PAC-Bayesian approach, and a number of results have been independently discovered in both strands. In this monograph, we highlight this strong connection and present a unified treatment of PAC-Bayesian and information-theoretic generalization bounds. We present techniques and results that the two perspectives have in common, and discuss the approaches and interpretations that differ. In particular, we demonstrate how many proofs in the area share a modular structure, through which the underlying ideas can be intuited. We pay special attention to the conditional mutual information (CMI) framework; analytical studies of the information complexity of learning algorithms; and the application of the proposed methods to deep learning. This monograph is intended to provide a comprehensive introduction to information-theoretic generalization bounds and their connection to PAC-Bayes, serving as a foundation from which the most recent developments are accessible. It is aimed broadly towards researchers with an interest in generalization and theoretical machine learning.
comment: 228 pages
♻ ☆ Decoupled Data Consistency with Diffusion Purification for Image Restoration
Diffusion models have recently gained traction as a powerful class of deep generative priors, excelling in a wide range of image restoration tasks due to their exceptional ability to model data distributions. To solve image restoration problems, many existing techniques achieve data consistency by incorporating additional likelihood gradient steps into the reverse sampling process of diffusion models. However, the additional gradient steps pose a challenge for real-world practical applications as they incur a large computational overhead, thereby increasing inference time. They also present additional difficulties when using accelerated diffusion model samplers, as the number of data consistency steps is limited by the number of reverse sampling steps. In this work, we propose a novel diffusion-based image restoration solver that addresses these issues by decoupling the reverse process from the data consistency steps. Our method involves alternating between a reconstruction phase to maintain data consistency and a refinement phase that enforces the prior via diffusion purification. Our approach demonstrates versatility, making it highly adaptable for efficient problem-solving in latent space. Additionally, it reduces the necessity for numerous sampling steps through the integration of consistency models. The efficacy of our approach is validated through comprehensive experiments across various image restoration tasks, including image denoising, deblurring, inpainting, and super-resolution.
♻ ☆ FedSN: A Novel Federated Learning Framework over LEO Satellite Networks
Recently, a large number of Low Earth Orbit (LEO) satellites have been launched and deployed successfully in space by commercial companies, such as SpaceX. Due to multimodal sensors equipped by the LEO satellites, they serve not only for communication but also for various machine learning applications, such as space modulation recognition, remote sensing image classification, etc. However, the ground station (GS) may be incapable of downloading such a large volume of raw sensing data for centralized model training due to the limited contact time with LEO satellites (e.g. 5 minutes). Therefore, federated learning (FL) has emerged as the promising solution to address this problem via on-device training. Unfortunately, to enable FL on LEO satellites, we still face three critical challenges that are i) heterogeneous computing and memory capabilities, ii) limited uplink rate, and iii) model staleness. To this end, we propose FedSN as a general FL framework to tackle the above challenges, and fully explore data diversity on LEO satellites. Specifically, we first present a novel sub-structure scheme to enable heterogeneous local model training considering different computing, memory, and communication constraints on LEO satellites. Additionally, we propose a pseudo-synchronous model aggregation strategy to dynamically schedule model aggregation for compensating model staleness. To further demonstrate the effectiveness of the FedSN, we evaluate it using space modulation recognition and remote sensing image classification tasks by leveraging the data from real-world satellite networks. Extensive experimental results demonstrate that FedSN framework achieves higher accuracy, lower computing, and communication overhead than the state-of-the-art benchmarks and the effectiveness of each components in FedSN.
comment: 14 pages, 17 figures
♻ ☆ Nonlinear Control Allocation: A Learning Based Approach IEEE Conference on Decision and Control
Modern aircraft are designed with redundant control effectors to cater for fault tolerance and maneuverability requirements. This leads to aircraft being over-actuated and requires control allocation schemes to distribute the control commands among control effectors. Traditionally, optimization-based control allocation schemes are used; however, for nonlinear allocation problems, these methods require large computational resources. In this work, an artificial neural network (ANN) based nonlinear control allocation scheme is proposed. The proposed scheme is composed of learning the inverse of the control effectiveness map through ANN, and then implementing it as an allocator instead of solving an online optimization problem. Stability conditions are presented for closed-loop systems incorporating the allocator, and computational challenges are explored with piece-wise linear effectiveness functions and ANN-based allocators. To demonstrate the efficacy of the proposed scheme, it is compared with a standard quadratic programming-based method for control allocation.
comment: submitted to IEEE Conference on Decision and Control (CDC), 2024
♻ ☆ NaturalSpeech 3: Zero-Shot Speech Synthesis with Factorized Codec and Diffusion Models
While recent large-scale text-to-speech (TTS) models have achieved significant progress, they still fall short in speech quality, similarity, and prosody. Considering speech intricately encompasses various attributes (e.g., content, prosody, timbre, and acoustic details) that pose significant challenges for generation, a natural idea is to factorize speech into individual subspaces representing different attributes and generate them individually. Motivated by it, we propose NaturalSpeech 3, a TTS system with novel factorized diffusion models to generate natural speech in a zero-shot way. Specifically, 1) we design a neural codec with factorized vector quantization (FVQ) to disentangle speech waveform into subspaces of content, prosody, timbre, and acoustic details; 2) we propose a factorized diffusion model to generate attributes in each subspace following its corresponding prompt. With this factorization design, NaturalSpeech 3 can effectively and efficiently model intricate speech with disentangled subspaces in a divide-and-conquer way. Experiments show that NaturalSpeech 3 outperforms the state-of-the-art TTS systems on quality, similarity, prosody, and intelligibility, and achieves on-par quality with human recordings. Furthermore, we achieve better performance by scaling to 1B parameters and 200K hours of training data.
comment: Achieving human-level quality and naturalness on multi-speaker datasets (e.g., LibriSpeech) in a zero-shot way
♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review
ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for EU AI policy act concerning ethics, digital divide, and sustainability.
comment: 29 pages, 8 figures, 4 tables
♻ ☆ Incorporating simulated spatial context information improves the effectiveness of contrastive learning models
Visual learning often occurs in a specific context, where an agent acquires skills through exploration and tracking of its location in a consistent environment. The historical spatial context of the agent provides a similarity signal for self-supervised contrastive learning. We present a unique approach, termed Environmental Spatial Similarity (ESS), that complements existing contrastive learning methods. Using images from simulated, photorealistic environments as an experimental setting, we demonstrate that ESS outperforms traditional instance discrimination approaches. Moreover, sampling additional data from the same environment substantially improves accuracy and provides new augmentations. ESS allows remarkable proficiency in room classification and spatial prediction tasks, especially in unfamiliar environments. This learning paradigm has the potential to enable rapid visual learning in agents operating in new environments with unique visual characteristics. Potentially transformative applications span from robotics to space exploration. Our proof of concept demonstrates improved efficiency over methods that rely on extensive, disconnected datasets.
♻ ☆ Stochastic Approximation with Delayed Updates: Finite-Time Rates under Markovian Sampling AISTATS
Motivated by applications in large-scale and multi-agent reinforcement learning, we study the non-asymptotic performance of stochastic approximation (SA) schemes with delayed updates under Markovian sampling. While the effect of delays has been extensively studied for optimization, the manner in which they interact with the underlying Markov process to shape the finite-time performance of SA remains poorly understood. In this context, our first main contribution is to show that under time-varying bounded delays, the delayed SA update rule guarantees exponentially fast convergence of the \emph{last iterate} to a ball around the SA operator's fixed point. Notably, our bound is \emph{tight} in its dependence on both the maximum delay $\tau_{max}$, and the mixing time $\tau_{mix}$. To achieve this tight bound, we develop a novel inductive proof technique that, unlike various existing delayed-optimization analyses, relies on establishing uniform boundedness of the iterates. As such, our proof may be of independent interest. Next, to mitigate the impact of the maximum delay on the convergence rate, we provide the first finite-time analysis of a delay-adaptive SA scheme under Markovian sampling. In particular, we show that the exponent of convergence of this scheme gets scaled down by $\tau_{avg}$, as opposed to $\tau_{max}$ for the vanilla delayed SA rule; here, $\tau_{avg}$ denotes the average delay across all iterations. Moreover, the adaptive scheme requires no prior knowledge of the delay sequence for step-size tuning. Our theoretical findings shed light on the finite-time effects of delays for a broad class of algorithms, including TD learning, Q-learning, and stochastic gradient descent under Markovian sampling.
comment: Accepted to the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024!
♻ ☆ Adaptive Negative Evidential Deep Learning for Open-set Semi-supervised Learning AAAI2024
Semi-supervised learning (SSL) methods assume that labeled data, unlabeled data and test data are from the same distribution. Open-set semi-supervised learning (Open-set SSL) considers a more practical scenario, where unlabeled data and test data contain new categories (outliers) not observed in labeled data (inliers). Most previous works focused on outlier detection via binary classifiers, which suffer from insufficient scalability and inability to distinguish different types of uncertainty. In this paper, we propose a novel framework, Adaptive Negative Evidential Deep Learning (ANEDL) to tackle these limitations. Concretely, we first introduce evidential deep learning (EDL) as an outlier detector to quantify different types of uncertainty, and design different uncertainty metrics for self-training and inference. Furthermore, we propose a novel adaptive negative optimization strategy, making EDL more tailored to the unlabeled dataset containing both inliers and outliers. As demonstrated empirically, our proposed method outperforms existing state-of-the-art methods across four datasets.
comment: Accepted by AAAI2024
♻ ☆ SeSaMe: A Framework to Simulate Self-Reported Ground Truth for Mental Health Sensing Studies
Advances in mobile and wearable technologies have enabled the potential to passively monitor a person's mental, behavioral, and affective health. These approaches typically rely on longitudinal collection of self-reported outcomes, e.g., depression, stress, and anxiety, to train machine learning (ML) models. However, the need to continuously self-report adds a significant burden on the participants, often resulting in attrition, missing labels, or insincere responses. In this work, we introduce the Scale Scores Simulation using Mental Models (SeSaMe) framework to alleviate participants' burden in digital mental health studies. By leveraging pre-trained large language models (LLMs), SeSaMe enables the simulation of participants' responses on psychological scales. In SeSaMe, researchers can prompt LLMs with information on participants' internal behavioral dispositions, enabling LLMs to construct mental models of participants to simulate their responses on psychological scales. We demonstrate an application of SeSaMe, where we use GPT-4 to simulate responses on one scale using responses from another as behavioral information. We also evaluate the alignment between human and SeSaMe-simulated responses to psychological scales. Then, we present experiments to inspect the utility of SeSaMe-simulated responses as ground truth in training ML models by replicating established depression and anxiety screening tasks from a previous study. Our results indicate SeSaMe to be a promising approach, but its alignment may vary across scales and specific prediction objectives. We also observed that model performance with simulated data was on par with using the real data for training in most evaluation scenarios. We conclude by discussing the potential implications of SeSaMe in addressing some challenges researchers face with ground-truth collection in passive sensing studies.
♻ ☆ Byzantine-resilient Federated Learning With Adaptivity to Data Heterogeneity
This paper deals with federated learning (FL) in the presence of malicious Byzantine attacks and data heterogeneity. A novel Robust Average Gradient Algorithm (RAGA) is proposed, which leverages the geometric median for aggregation and can freely select the round number for local updating. Different from most existing resilient approaches, which perform convergence analysis based on strongly-convex loss function or homogeneously distributed dataset, we conduct convergence analysis for not only strongly-convex but also non-convex loss function over heterogeneous dataset. According to our theoretical analysis, as long as the fraction of dataset from malicious users is less than half, RAGA can achieve convergence at rate $\mathcal{O}({1}/{T^{2/3- \delta}})$ where $T$ is the iteration number and $\delta \in (0, 2/3)$ for non-convex loss function, and at linear rate for strongly-convex loss function. Moreover, stationary point or global optimal solution is proved to obtainable as data heterogeneity vanishes. Experimental results corroborate the robustness of RAGA to Byzantine attacks and verifies the advantage of RAGA over baselines on convergence performance under various intensity of Byzantine attacks, for heterogeneous dataset.
♻ ☆ Demystifying Misconceptions in Social Bots Research
Research on social bots aims at advancing knowledge and providing solutions to one of the most debated forms of online manipulation. Yet, social bot research is plagued by widespread biases, hyped results, and misconceptions that set the stage for ambiguities, unrealistic expectations, and seemingly irreconcilable findings. Overcoming such issues is instrumental towards ensuring reliable solutions and reaffirming the validity of the scientific method. In this contribution, we review some recent results in social bots research, highlighting and revising factual errors as well as methodological and conceptual biases. More importantly, we demystify common misconceptions, addressing fundamental points on how social bots research is discussed. Our analysis surfaces the need to discuss research about online disinformation and manipulation in a rigorous, unbiased, and responsible way. This article bolsters such effort by identifying and refuting common fallacious arguments used by both proponents and opponents of social bots research, as well as providing directions toward sound methodologies for future research in the field.
♻ ☆ DeepMachining: Online Prediction of Machining Errors of Lathe Machines
We describe DeepMachining, a deep learning-based AI system for online prediction of machining errors of lathe machine operations. We have built and evaluated DeepMachining based on manufacturing data from factories. Specifically, we first pretrain a deep learning model for a given lathe machine's operations to learn the salient features of machining states. Then, we fine-tune the pretrained model to adapt to specific machining tasks. We demonstrate that DeepMachining achieves high prediction accuracy for multiple tasks that involve different workpieces and cutting tools. To the best of our knowledge, this work is one of the first factory experiments using pre-trained deep-learning models to predict machining errors of lathe machines.
♻ ☆ Structure Guided Large Language Model for SQL Generation
Generating accurate Structured Querying Language (SQL) is a long-standing problem, especially in matching users' semantic queries with structured databases and then generating structured SQL. Existing models typically input queries and database schemas into the LLM and rely on the LLM to perform semantic-structure matching and generate structured SQL. However, such solutions overlook the structural information within user queries and databases, which can be utilized to enhance the generation of structured SQL. This oversight can lead to inaccurate or unexecutable SQL generation. To fully exploit the structure, we propose a structure-to-SQL framework, which leverages the inherent structure information to improve the SQL generation of LLMs. Specifically, we introduce our Structure Guided SQL~(SGU-SQL) generation model. SGU-SQL first links user queries and databases in a structure-enhanced manner. It then decomposes complicated linked structures with grammar trees to guide the LLM to generate the SQL step by step. Extensive experiments on two benchmark datasets illustrate that SGU-SQL can outperform sixteen SQL generation baselines.
♻ ☆ Recurrent Action Transformer with Memory
Recently, the use of transformers in offline reinforcement learning has become a rapidly developing area. This is due to their ability to treat the agent's trajectory in the environment as a sequence, thereby reducing the policy learning problem to sequence modeling. In environments where the agent's decisions depend on past events, it is essential to capture both the event itself and the decision point in the context of the model. However, the quadratic complexity of the attention mechanism limits the potential for context expansion. One solution to this problem is to enhance transformers with memory mechanisms. In this paper, we propose the Recurrent Action Transformer with Memory (RATE) - a model that incorporates recurrent memory. To evaluate our model, we conducted extensive experiments on both memory-intensive environments (VizDoom-Two-Color, T-Maze) and classic Atari games and MuJoCo control environments. The results show that the use of memory can significantly improve performance in memory-intensive environments while maintaining or improving results in classic environments. We hope that our findings will stimulate research on memory mechanisms for transformers applicable to offline reinforcement learning.
comment: 15 pages, 11 figures
♻ ☆ Generative Pre-Training of Time-Series Data for Unsupervised Fault Detection in Semiconductor Manufacturing
This paper introduces TRACE-GPT, which stands for Time-seRies Anomaly-detection with Convolutional Embedding and Generative Pre-trained Transformers. TRACE-GPT is designed to pre-train univariate time-series sensor data and detect faults on unlabeled datasets in semiconductor manufacturing. In semiconductor industry, classifying abnormal time-series sensor data from normal data is important because it is directly related to wafer defect. However, small, unlabeled, and even mixed training data without enough anomalies make classification tasks difficult. In this research, we capture features of time-series data with temporal convolutional embedding and Generative Pre-trained Transformer (GPT) to classify abnormal sequences from normal sequences using cross entropy loss. We prove that our model shows better performance than previous unsupervised models with both an open dataset, the University of California Riverside (UCR) time-series classification archive, and the process log of our Chemical Vapor Deposition (CVD) equipment. Our model has the highest F1 score at Equal Error Rate (EER) across all datasets and is only 0.026 below the supervised state-of-the-art baseline on the open dataset.
♻ ☆ Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey NAACL 2024
Large Language Models (LLMs) are now commonplace in conversation applications. However, their risks of misuse for generating harmful responses have raised serious societal concerns and spurred recent research on LLM conversation safety. Therefore, in this survey, we provide a comprehensive overview of recent studies, covering three critical aspects of LLM conversation safety: attacks, defenses, and evaluations. Our goal is to provide a structured summary that enhances understanding of LLM conversation safety and encourages further investigation into this important subject. For easy reference, we have categorized all the studies mentioned in this survey according to our taxonomy, available at: https://github.com/niconi19/LLM-conversation-safety.
comment: Accepted to NAACL 2024
♻ ☆ Shapley Values-Powered Framework for Fair Reward Split in Content Produced by GenAI
It is evident that, currently, generative models are surpassed in quality by human professionals. However, with the advancements in Artificial Intelligence, this gap will narrow, leading to scenarios where individuals who have dedicated years of their lives to mastering a skill become obsolete due to their high costs, which are inherently linked to the time they require to complete a task -- a task that AI could accomplish in minutes or seconds. To avoid future social upheavals, we must, even now, contemplate how to fairly assess the contributions of such individuals in training generative models and how to compensate them for the reduction or complete loss of their incomes. In this work, we propose a method to structure collaboration between model developers and data providers. To achieve this, we employ Shapley Values to quantify the contribution of artist(s) in an image generated by the Stable Diffusion-v1.5 model and to equitably allocate the reward among them.
comment: 36 pages, 32 figures
♻ ☆ ABScribe: Rapid Exploration & Organization of Multiple Writing Variations in Human-AI Co-Writing Tasks using Large Language Models
Exploring alternative ideas by rewriting text is integral to the writing process. State-of-the-art Large Language Models (LLMs) can simplify writing variation generation. However, current interfaces pose challenges for simultaneous consideration of multiple variations: creating new variations without overwriting text can be difficult, and pasting them sequentially can clutter documents, increasing workload and disrupting writers' flow. To tackle this, we present ABScribe, an interface that supports rapid, yet visually structured, exploration and organization of writing variations in human-AI co-writing tasks. With ABScribe, users can swiftly modify variations using LLM prompts, which are auto-converted into reusable buttons. Variations are stored adjacently within text fields for rapid in-place comparisons using mouse-over interactions on a popup toolbar. Our user study with 12 writers shows that ABScribe significantly reduces task workload (d = 1.20, p < 0.001), enhances user perceptions of the revision process (d = 2.41, p < 0.001) compared to a popular baseline workflow, and provides insights into how writers explore variations using LLMs.
comment: CHI 2024
♻ ☆ CroSel: Cross Selection of Confident Pseudo Labels for Partial-Label Learning CVPR 2024
Partial-label learning (PLL) is an important weakly supervised learning problem, which allows each training example to have a candidate label set instead of a single ground-truth label. Identification-based methods have been widely explored to tackle label ambiguity issues in PLL, which regard the true label as a latent variable to be identified. However, identifying the true labels accurately and completely remains challenging, causing noise in pseudo labels during model training. In this paper, we propose a new method called CroSel, which leverages historical predictions from the model to identify true labels for most training examples. First, we introduce a cross selection strategy, which enables two deep models to select true labels of partially labeled data for each other. Besides, we propose a novel consistency regularization term called co-mix to avoid sample waste and tiny noise caused by false selection. In this way, CroSel can pick out the true labels of most examples with high precision. Extensive experiments demonstrate the superiority of CroSel, which consistently outperforms previous state-of-the-art methods on benchmark datasets. Additionally, our method achieves over 90\% accuracy and quantity for selecting true labels on CIFAR-type datasets under various settings.
comment: Accepted by CVPR 2024
♻ ☆ Challenging Common Paradigms in Multi-Task Learning
While multi-task learning (MTL) has gained significant attention in recent years, its underlying mechanisms remain poorly understood. Recent methods did not yield consistent performance improvements over single task learning (STL) baselines, underscoring the importance of gaining more profound insights about challenges specific to MTL. In our study, we challenge paradigms in MTL in the context of STL: First, the impact of the choice of optimizer has only been mildly investigated in MTL. We show the pivotal role of common STL tools such as the Adam optimizer in MTL empirically in various experiments. To further investigate Adam's effectiveness, we theoretical derive a partial loss-scale invariance under mild assumptions. Second, the notion of gradient conflicts has often been phrased as a specific problem in MTL. We delve into the role of gradient conflicts in MTL and compare it to STL. For angular gradient alignment we find no evidence that this is a unique problem in MTL. We emphasize differences in gradient magnitude as the main distinguishing factor. Lastly, we compare the transferability of features learned through MTL and STL on common image corruptions, and find light evidence that MTL can lead to superior transferability. Overall, we find surprising similarities between STL and MTL suggesting to consider methods from both fields in a broader context.
comment: -
♻ ☆ Solving a Real-World Package Delivery Routing Problem Using Quantum Annealers
Research focused on the conjunction between quantum computing and routing problems has been very prolific in recent years. Most of the works revolve around classical problems such as the Traveling Salesman Problem or the Vehicle Routing Problem. Even though working on these problems is valuable, it is also undeniable that their academic-oriented nature falls short of real-world requirements. The main objective of this research is to present a solving method for realistic instances, avoiding problem relaxations or technical shortcuts. Instead, a quantum-classical hybrid solver has been developed, coined Q4RPD, that considers a set of real constraints such as a heterogeneous fleet of vehicles, priority deliveries, and capacities characterized by two values: weight and dimensions of the packages. Q4RPD resorts to the Leap Constrained Quadratic Model Hybrid Solver of D-Wave. To demonstrate the application of Q4RPD, an experimentation composed of six different instances has been conducted, aiming to serve as illustrative examples.
comment: 15 pages, 11 figures and 4 tables. Paper submitted for review in Scientific Reports
♻ ☆ Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation CVPR 2024
Transformers have been successfully applied in the field of video-based 3D human pose estimation. However, the high computational costs of these video pose transformers (VPTs) make them impractical on resource-constrained devices. In this paper, we present a plug-and-play pruning-and-recovering framework, called Hourglass Tokenizer (HoT), for efficient transformer-based 3D human pose estimation from videos. Our HoT begins with pruning pose tokens of redundant frames and ends with recovering full-length tokens, resulting in a few pose tokens in the intermediate transformer blocks and thus improving the model efficiency. To effectively achieve this, we propose a token pruning cluster (TPC) that dynamically selects a few representative tokens with high semantic diversity while eliminating the redundancy of video frames. In addition, we develop a token recovering attention (TRA) to restore the detailed spatio-temporal information based on the selected tokens, thereby expanding the network output to the original full-length temporal resolution for fast inference. Extensive experiments on two benchmark datasets (i.e., Human3.6M and MPI-INF-3DHP) demonstrate that our method can achieve both high efficiency and estimation accuracy compared to the original VPT models. For instance, applying to MotionBERT and MixSTE on Human3.6M, our HoT can save nearly 50% FLOPs without sacrificing accuracy and nearly 40% FLOPs with only 0.2% accuracy drop, respectively. Code and models are available at https://github.com/NationalGAILab/HoT.
comment: Accepted by CVPR 2024, Open Sourced
♻ ☆ $\textit{LinkPrompt}$: Natural and Universal Adversarial Attacks on Prompt-based Language Models NAACL2024
Prompt-based learning is a new language model training paradigm that adapts the Pre-trained Language Models (PLMs) to downstream tasks, which revitalizes the performance benchmarks across various natural language processing (NLP) tasks. Instead of using a fixed prompt template to fine-tune the model, some research demonstrates the effectiveness of searching for the prompt via optimization. Such prompt optimization process of prompt-based learning on PLMs also gives insight into generating adversarial prompts to mislead the model, raising concerns about the adversarial vulnerability of this paradigm. Recent studies have shown that universal adversarial triggers (UATs) can be generated to alter not only the predictions of the target PLMs but also the prediction of corresponding Prompt-based Fine-tuning Models (PFMs) under the prompt-based learning paradigm. However, UATs found in previous works are often unreadable tokens or characters and can be easily distinguished from natural texts with adaptive defenses. In this work, we consider the naturalness of the UATs and develop $\textit{LinkPrompt}$, an adversarial attack algorithm to generate UATs by a gradient-based beam search algorithm that not only effectively attacks the target PLMs and PFMs but also maintains the naturalness among the trigger tokens. Extensive results demonstrate the effectiveness of $\textit{LinkPrompt}$, as well as the transferability of UATs generated by $\textit{LinkPrompt}$ to open-sourced Large Language Model (LLM) Llama2 and API-accessed LLM GPT-3.5-turbo.
comment: Accepted to the main conference of NAACL2024
♻ ☆ LLatrieval: LLM-Verified Retrieval for Verifiable Generation NAACL 2024
Verifiable generation aims to let the large language model (LLM) generate text with supporting documents, which enables the user to flexibly verify the answer and makes the LLM's output more reliable. Retrieval plays a crucial role in verifiable generation. Specifically, the retrieved documents not only supplement knowledge to help the LLM generate correct answers, but also serve as supporting evidence for the user to verify the LLM's output. However, the widely used retrievers become the bottleneck of the entire pipeline and limit the overall performance. Their capabilities are usually inferior to LLMs since they often have much fewer parameters than the large language model and have not been demonstrated to scale well to the size of LLMs. If the retriever does not correctly find the supporting documents, the LLM can not generate the correct and verifiable answer, which overshadows the LLM's remarkable abilities. To address these limitations, we propose \LLatrieval (Large Language Model Verified Retrieval), where the LLM updates the retrieval result until it verifies that the retrieved documents can sufficiently support answering the question. Thus, the LLM can iteratively provide feedback to retrieval and facilitate the retrieval result to fully support verifiable generation. Experiments show that LLatrieval significantly outperforms extensive baselines and achieves state-of-the-art results.
comment: Accepted by NAACL 2024 (Main Conference)
♻ ☆ Enhancing Object Coherence in Layout-to-Image Synthesis
Layout-to-image synthesis is an emerging technique in conditional image generation. It aims to generate complex scenes, where users require fine control over the layout of the objects in a scene. However, it remains challenging to control the object coherence, including semantic coherence (e.g., the cat looks at the flowers or not) and physical coherence (e.g., the hand and the racket should not be misaligned). In this paper, we propose a novel diffusion model with effective global semantic fusion (GSF) and self-similarity feature enhancement modules to guide the object coherence for this task. For semantic coherence, we argue that the image caption contains rich information for defining the semantic relationship within the objects in the images. Instead of simply employing cross-attention between captions and generated images, which addresses the highly relevant layout restriction and semantic coherence separately and thus leads to unsatisfying results shown in our experiments, we develop GSF to fuse the supervision from the layout restriction and semantic coherence requirement and exploit it to guide the image synthesis process. Moreover, to improve the physical coherence, we develop a Self-similarity Coherence Attention (SCA) module to explicitly integrate local contextual physical coherence into each pixel's generation process. Specifically, we adopt a self-similarity map to encode the coherence restrictions and employ it to extract coherent features from text embedding. Through visualization of our self-similarity map, we explore the essence of SCA, revealing that its effectiveness is not only in capturing reliable physical coherence patterns but also in enhancing complex texture generation. Extensive experiments demonstrate the superiority of our proposed method in both image generation quality and controllability.
♻ ☆ OpenMoE: An Early Effort on Open Mixture-of-Experts Language Models
To help the open-source community have a better understanding of Mixture-of-Experts (MoE) based large language models (LLMs), we train and release OpenMoE, a series of fully open-sourced and reproducible decoder-only MoE LLMs, ranging from 650M to 34B parameters and trained on up to over 1T tokens. Our investigation confirms that MoE-based LLMs can offer a more favorable cost-effectiveness trade-off than dense LLMs, highlighting the potential effectiveness for future LLM development. One more important contribution of this study is an in-depth analysis of the routing mechanisms within our OpenMoE models, leading to three significant findings: Context-Independent Specialization, Early Routing Learning, and Drop-towards-the-End. We discovered that routing decisions in MoE models are predominantly based on token IDs, with minimal context relevance. The token-to-expert assignments are determined early in the pre-training phase and remain largely unchanged. This imperfect routing can result in performance degradation, particularly in sequential tasks like multi-turn conversations, where tokens appearing later in a sequence are more likely to be dropped. Finally, we rethink our design based on the above-mentioned observations and analysis. To facilitate future MoE LLM development, we propose potential strategies for mitigating the issues we found and further improving off-the-shelf MoE LLM designs.
VIGraph: Generative Self-supervised Learning for Class-Imbalanced Node Classification
Class imbalance in graph data presents significant challenges for node classification. While existing methods, such as SMOTE-based approaches, partially mitigate this issue, they still exhibit limitations in constructing imbalanced graphs. Generative self-supervised learning (SSL) methods, exemplified by graph autoencoders (GAEs), offer a promising solution by directly generating minority nodes from the data itself, yet their potential remains underexplored. In this paper, we delve into the shortcomings of SMOTE-based approaches in the construction of imbalanced graphs. Furthermore, we introduce VIGraph, a simple yet effective generative SSL approach that relies on the Variational GAE as the fundamental model. VIGraph strictly adheres to the concept of imbalance when constructing imbalanced graphs and innovatively leverages the variational inference (VI) ability of Variational GAE to generate nodes for minority classes. VIGraph introduces comprehensive training strategies, including cross-view contrastive learning at the decoding phase to capture semantic knowledge, adjacency matrix reconstruction to preserve graph structure, and alignment strategy to ensure stable training. VIGraph can generate high-quality nodes directly usable for classification, eliminating the need to integrate the generated nodes back to the graph as well as additional retraining found in SMOTE-based methods. We conduct extensive experiments, results from which demonstrate the superiority and generality of our approach.
Learning Quadruped Locomotion Using Differentiable Simulation
While most recent advancements in legged robot control have been driven by model-free reinforcement learning, we explore the potential of differentiable simulation. Differentiable simulation promises faster convergence and more stable training by computing low-variant first-order gradients using the robot model, but so far, its use for legged robot control has remained limited to simulation. The main challenge with differentiable simulation lies in the complex optimization landscape of robotic tasks due to discontinuities in contact-rich environments, e.g., quadruped locomotion. This work proposes a new, differentiable simulation framework to overcome these challenges. The key idea involves decoupling the complex whole-body simulation, which may exhibit discontinuities due to contact, into two separate continuous domains. Subsequently, we align the robot state resulting from the simplified model with a more precise, non-differentiable simulator to maintain sufficient simulation accuracy. Our framework enables learning quadruped walking in minutes using a single simulated robot without any parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills, including trot, pace, bound, and gallop, on challenging terrains in minutes. Additionally, our policy achieves robust locomotion performance in the real world zero-shot. To the best of our knowledge, this work represents the first demonstration of using differentiable simulation for controlling a real quadruped robot. This work provides several important insights into using differentiable simulations for legged locomotion in the real world.
♻ ☆ Retrieval-Augmented Generation for Large Language Models: A Survey
Large Language Models (LLMs) showcase impressive capabilities but encounter challenges like hallucination, outdated knowledge, and non-transparent, untraceable reasoning processes. Retrieval-Augmented Generation (RAG) has emerged as a promising solution by incorporating knowledge from external databases. This enhances the accuracy and credibility of the generation, particularly for knowledge-intensive tasks, and allows for continuous knowledge updates and integration of domain-specific information. RAG synergistically merges LLMs' intrinsic knowledge with the vast, dynamic repositories of external databases. This comprehensive review paper offers a detailed examination of the progression of RAG paradigms, encompassing the Naive RAG, the Advanced RAG, and the Modular RAG. It meticulously scrutinizes the tripartite foundation of RAG frameworks, which includes the retrieval, the generation and the augmentation techniques. The paper highlights the state-of-the-art technologies embedded in each of these critical components, providing a profound understanding of the advancements in RAG systems. Furthermore, this paper introduces up-to-date evaluation framework and benchmark. At the end, this article delineates the challenges currently faced and points out prospective avenues for research and development.
comment: Ongoing Work
♻ ☆ World Models via Policy-Guided Trajectory Diffusion
World models are a powerful tool for developing intelligent agents. By predicting the outcome of a sequence of actions, world models enable policies to be optimised via on-policy reinforcement learning (RL) using synthetic data, i.e. in "in imagination". Existing world models are autoregressive in that they interleave predicting the next state with sampling the next action from the policy. Prediction error inevitably compounds as the trajectory length grows. In this work, we propose a novel world modelling approach that is not autoregressive and generates entire on-policy trajectories in a single pass through a diffusion model. Our approach, Policy-Guided Trajectory Diffusion (PolyGRAD), leverages a denoising model in addition to the gradient of the action distribution of the policy to diffuse a trajectory of initially random states and actions into an on-policy synthetic trajectory. We analyse the connections between PolyGRAD, score-based generative models, and classifier-guided diffusion models. Our results demonstrate that PolyGRAD outperforms state-of-the-art baselines in terms of trajectory prediction error for short trajectories, with the exception of autoregressive diffusion. For short trajectories, PolyGRAD obtains similar errors to autoregressive diffusion, but with lower computational requirements. For long trajectories, PolyGRAD obtains comparable performance to baselines. Our experiments demonstrate that PolyGRAD enables performant policies to be trained via on-policy RL in imagination for MuJoCo continuous control domains. Thus, PolyGRAD introduces a new paradigm for accurate on-policy world modelling without autoregressive sampling.
comment: Published in TMLR, March 2024
♻ ☆ Functional Graph Convolutional Networks: A unified multi-task and multi-modal learning framework to facilitate health and social-care insights
This paper introduces a novel Functional Graph Convolutional Network (funGCN) framework that combines Functional Data Analysis and Graph Convolutional Networks to address the complexities of multi-task and multi-modal learning in digital health and longitudinal studies. With the growing importance of health solutions to improve health care and social support, ensure healthy lives, and promote well-being at all ages, funGCN offers a unified approach to handle multivariate longitudinal data for multiple entities and ensures interpretability even with small sample sizes. Key innovations include task-specific embedding components that manage different data types, the ability to perform classification, regression, and forecasting, and the creation of a knowledge graph for insightful data interpretation. The efficacy of funGCN is validated through simulation experiments and a real-data application.
♻ ☆ Learning Concept-Based Causal Transition and Symbolic Reasoning for Visual Planning
Visual planning simulates how humans make decisions to achieve desired goals in the form of searching for visual causal transitions between an initial visual state and a final visual goal state. It has become increasingly important in egocentric vision with its advantages in guiding agents to perform daily tasks in complex environments. In this paper, we propose an interpretable and generalizable visual planning framework consisting of i) a novel Substitution-based Concept Learner (SCL) that abstracts visual inputs into disentangled concept representations, ii) symbol abstraction and reasoning that performs task planning via the self-learned symbols, and iii) a Visual Causal Transition model (ViCT) that grounds visual causal transitions to semantically similar real-world actions. Given an initial state, we perform goal-conditioned visual planning with a symbolic reasoning method fueled by the learned representations and causal transitions to reach the goal state. To verify the effectiveness of the proposed model, we collect a large-scale visual planning dataset based on AI2-THOR, dubbed as CCTP. Extensive experiments on this challenging dataset demonstrate the superior performance of our method in visual task planning. Empirically, we show that our framework can generalize to unseen task trajectories, unseen object categories, and real-world data. Further details of this work are provided at https://fqyqc.github.io/ConTranPlan/.
♻ ☆ Identifying the Correlation Between Language Distance and Cross-Lingual Transfer in a Multilingual Representation Space EACL 2023
Prior research has investigated the impact of various linguistic features on cross-lingual transfer performance. In this study, we investigate the manner in which this effect can be mapped onto the representation space. While past studies have focused on the impact on cross-lingual alignment in multilingual language models during fine-tuning, this study examines the absolute evolution of the respective language representation spaces produced by MLLMs. We place a specific emphasis on the role of linguistic characteristics and investigate their inter-correlation with the impact on representation spaces and cross-lingual transfer performance. Additionally, this paper provides preliminary evidence of how these findings can be leveraged to enhance transfer to linguistically distant languages.
comment: SIGTYP Workshop 2023 (co-located with EACL 2023)
♻ ☆ The Effects of Mixed Sample Data Augmentation are Class Dependent
Mixed Sample Data Augmentation (MSDA) techniques, such as Mixup, CutMix, and PuzzleMix, have been widely acknowledged for enhancing performance in a variety of tasks. A previous study reported the class dependency of traditional data augmentation (DA), where certain classes benefit disproportionately compared to others. This paper reveals a class dependent effect of MSDA, where some classes experience improved performance while others experience degraded performance. This research addresses the issue of class dependency in MSDA and proposes an algorithm to mitigate it. The approach involves training on a mixture of MSDA and non-MSDA data, which not only mitigates the negative impact on the affected classes, but also improves overall accuracy. Furthermore, we provide in-depth analysis and discussion of why MSDA introduced class dependencies and which classes are most likely to have them.
comment: 21 pages, 18 figures, Overall Revision
♻ ☆ Spectral Meets Spatial: Harmonising 3D Shape Matching and Interpolation CVPR2024
Although 3D shape matching and interpolation are highly interrelated, they are often studied separately and applied sequentially to relate different 3D shapes, thus resulting in sub-optimal performance. In this work we present a unified framework to predict both point-wise correspondences and shape interpolation between 3D shapes. To this end, we combine the deep functional map framework with classical surface deformation models to map shapes in both spectral and spatial domains. On the one hand, by incorporating spatial maps, our method obtains more accurate and smooth point-wise correspondences compared to previous functional map methods for shape matching. On the other hand, by introducing spectral maps, our method gets rid of commonly used but computationally expensive geodesic distance constraints that are only valid for near-isometric shape deformations. Furthermore, we propose a novel test-time adaptation scheme to capture both pose-dominant and shape-dominant deformations. Using different challenging datasets, we demonstrate that our method outperforms previous state-of-the-art methods for both shape matching and interpolation, even compared to supervised approaches.
comment: accepted by CVPR2024
♻ ☆ Towards Regulatable AI Systems: Technical Gaps and Policy Opportunities
There is increasing attention being given to how to regulate AI systems. As governing bodies grapple with what values to encapsulate into regulation, we consider the technical half of the question: To what extent can AI experts vet an AI system for adherence to regulatory requirements? We investigate this question through the lens of two public sector procurement checklists, identifying what we can do now, what should be possible with technical innovation, and what requirements need a more interdisciplinary approach.
comment: scheduled for publication in the Communications of the ACM, titled "Directions of Technical Innovation for Regulatable AI Systems"
♻ ☆ MMP++: Motion Manifold Primitives with Parametric Curve Models
Motion Manifold Primitives (MMP), a manifold-based approach for encoding basic motion skills, can produce diverse trajectories, enabling the system to adapt to unseen constraints. Nonetheless, we argue that current MMP models lack crucial functionalities of movement primitives, such as temporal and via-points modulation, found in traditional approaches. This shortfall primarily stems from MMP's reliance on discrete-time trajectories. To overcome these limitations, we introduce Motion Manifold Primitives++ (MMP++), a new model that integrates the strengths of both MMP and traditional methods by incorporating parametric curve representations into the MMP framework. Furthermore, we identify a significant challenge with MMP++: performance degradation due to geometric distortions in the latent space, meaning that similar motions are not closely positioned. To address this, Isometric Motion Manifold Primitives++ (IMMP++) is proposed to ensure the latent space accurately preserves the manifold's geometry. Our experimental results across various applications, including 2-DoF planar motions, 7-DoF robot arm motions, and SE(3) trajectory planning, show that MMP++ and IMMP++ outperform existing methods in trajectory generation tasks, achieving substantial improvements in some cases. Moreover, they enable the modulation of latent coordinates and via-points, thereby allowing efficient online adaptation to dynamic environments.
comment: 12 pages. This work has been submitted to the IEEE for possible publication
♻ ☆ Regret-Based Defense in Adversarial Reinforcement Learning AAMAS 2024
Deep Reinforcement Learning (DRL) policies have been shown to be vulnerable to small adversarial noise in observations. Such adversarial noise can have disastrous consequences in safety-critical environments. For instance, a self-driving car receiving adversarially perturbed sensory observations about nearby signs (e.g., a stop sign physically altered to be perceived as a speed limit sign) or objects (e.g., cars altered to be recognized as trees) can be fatal. Existing approaches for making RL algorithms robust to an observation-perturbing adversary have focused on reactive approaches that iteratively improve against adversarial examples generated at each iteration. While such approaches have been shown to provide improvements over regular RL methods, they are reactive and can fare significantly worse if certain categories of adversarial examples are not generated during training. To that end, we pursue a more proactive approach that relies on directly optimizing a well-studied robustness measure, regret instead of expected value. We provide a principled approach that minimizes maximum regret over a "neighborhood" of observations to the received "observation". Our regret criterion can be used to modify existing value- and policy-based Deep RL methods. We demonstrate that our approaches provide a significant improvement in performance across a wide variety of benchmarks against leading approaches for robust Deep RL.
comment: Accepted at AAMAS 2024
♻ ☆ MA4DIV: Multi-Agent Reinforcement Learning for Search Result Diversification
The objective of search result diversification (SRD) is to ensure that selected documents cover as many different subtopics as possible. Existing methods primarily utilize a paradigm of "greedy selection", i.e., selecting one document with the highest diversity score at a time. These approaches tend to be inefficient and are easily trapped in a suboptimal state. In addition, some other methods aim to approximately optimize the diversity metric, such as $\alpha$-NDCG, but the results still remain suboptimal. To address these challenges, we introduce Multi-Agent reinforcement learning (MARL) for search result DIVersity, which called MA4DIV. In this approach, each document is an agent and the search result diversification is modeled as a cooperative task among multiple agents. This approach allows for directly optimizing the diversity metrics, such as $\alpha$-NDCG, while achieving high training efficiency. We conducted preliminary experiments on public TREC datasets to demonstrate the effectiveness and potential of MA4DIV. Considering the limited number of queries in public TREC datasets, we construct a large-scale dataset from industry sources and show that MA4DIV achieves substantial improvements in both effectiveness and efficiency than existing baselines on a industrial scale dataset.
♻ ☆ LLMs Are Few-Shot In-Context Low-Resource Language Learners
In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages.
♻ ☆ Enhancing Programming Education with ChatGPT: A Case Study on Student Perceptions and Interactions in a Python Course
The integration of ChatGPT as a supportive tool in education, notably in programming courses, addresses the unique challenges of programming education by providing assistance with debugging, code generation, and explanations. Despite existing research validating ChatGPT's effectiveness, its application in university-level programming education and a detailed understanding of student interactions and perspectives remain limited. This paper explores ChatGPT's impact on learning in a Python programming course tailored for first-year students over eight weeks. By analyzing responses from surveys, open-ended questions, and student-ChatGPT dialog data, we aim to provide a comprehensive view of ChatGPT's utility and identify both its advantages and limitations as perceived by students. Our study uncovers a generally positive reception toward ChatGPT and offers insights into its role in enhancing the programming education experience. These findings contribute to the broader discourse on AI's potential in education, suggesting paths for future research and application.
♻ ☆ SSM Meets Video Diffusion Models: Efficient Video Generation with Structured State Spaces ICLR 2024
Given the remarkable achievements in image generation through diffusion models, the research community has shown increasing interest in extending these models to video generation. Recent diffusion models for video generation have predominantly utilized attention layers to extract temporal features. However, attention layers are limited by their memory consumption, which increases quadratically with the length of the sequence. This limitation presents significant challenges when attempting to generate longer video sequences using diffusion models. To overcome this challenge, we propose leveraging state-space models (SSMs). SSMs have recently gained attention as viable alternatives due to their linear memory consumption relative to sequence length. In the experiments, we first evaluate our SSM-based model with UCF101, a standard benchmark of video generation. In addition, to investigate the potential of SSMs for longer video generation, we perform an experiment using the MineRL Navigate dataset, varying the number of frames to 64, 200, and 400. In these settings, our SSM-based model can considerably save memory consumption for longer sequences, while maintaining competitive FVD scores to the attention-based models. Our codes are available at https://github.com/shim0114/SSM-Meets-Video-Diffusion-Models.
comment: Accepted as workshop paper at ICLR 2024
♻ ☆ Weakly Supervised AUC Optimization: A Unified Partial AUC Approach
Since acquiring perfect supervision is usually difficult, real-world machine learning tasks often confront inaccurate, incomplete, or inexact supervision, collectively referred to as weak supervision. In this work, we present WSAUC, a unified framework for weakly supervised AUC optimization problems, which covers noisy label learning, positive-unlabeled learning, multi-instance learning, and semi-supervised learning scenarios. Within the WSAUC framework, we first frame the AUC optimization problems in various weakly supervised scenarios as a common formulation of minimizing the AUC risk on contaminated sets, and demonstrate that the empirical risk minimization problems are consistent with the true AUC. Then, we introduce a new type of partial AUC, specifically, the reversed partial AUC (rpAUC), which serves as a robust training objective for AUC maximization in the presence of contaminated labels. WSAUC offers a universal solution for AUC optimization in various weakly supervised scenarios by maximizing the empirical rpAUC. Theoretical and experimental results under multiple settings support the effectiveness of WSAUC on a range of weakly supervised AUC optimization tasks.
comment: Accepted by IEEE TPAMI
♻ ☆ ProSwitch: Knowledge-Guided Language Model Fine-Tuning to Generate Professional and Non-Professional Styled Text
Large Language Models (LLMs) have demonstrated efficacy in various linguistic applications, including text summarization and controlled text generation. However, studies into their capacity of switching between styles via fine-tuning remain underexplored. This study concentrates on textual professionalism and introduces a novel methodology, named ProSwitch, which equips a language model with the ability to produce both professional and non-professional responses through knowledge-guided instruction tuning. ProSwitch unfolds across three phases: data preparation for gathering domain knowledge and training corpus; instruction tuning for optimizing language models with multiple levels of instruction formats; and comprehensive evaluation for assessing the professionalism discrimination and reference-based quality of generated text. Comparative analysis of ProSwitch against both general and specialized language models reveals that our approach outperforms baselines in switching between professional and non-professional text generation.
comment: 8 pages
♻ ☆ Re2LLM: Reflective Reinforcement Large Language Model for Session-based Recommendation
Large Language Models (LLMs) are emerging as promising approaches to enhance session-based recommendation (SBR), where both prompt-based and fine-tuning-based methods have been widely investigated to align LLMs with SBR. However, the former methods struggle with optimal prompts to elicit the correct reasoning of LLMs due to the lack of task-specific feedback, leading to unsatisfactory recommendations. Although the latter methods attempt to fine-tune LLMs with domain-specific knowledge, they face limitations such as high computational costs and reliance on open-source backbones. To address such issues, we propose a Reflective Reinforcement Large Language Model (Re2LLM) for SBR, guiding LLMs to focus on specialized knowledge essential for more accurate recommendations effectively and efficiently. In particular, we first design the Reflective Exploration Module to effectively extract knowledge that is readily understandable and digestible by LLMs. To be specific, we direct LLMs to examine recommendation errors through self-reflection and construct a knowledge base (KB) comprising hints capable of rectifying these errors. To efficiently elicit the correct reasoning of LLMs, we further devise the Reinforcement Utilization Module to train a lightweight retrieval agent. It learns to select hints from the constructed KB based on the task-specific feedback, where the hints can serve as guidance to help correct LLMs reasoning for better recommendations. Extensive experiments on multiple real-world datasets demonstrate that our method consistently outperforms state-of-the-art methods.
comment: 11 pages, 4 figures
♻ ☆ Dial-MAE: ConTextual Masked Auto-Encoder for Retrieval-based Dialogue Systems NAACL 2024
Dialogue response selection aims to select an appropriate response from several candidates based on a given user and system utterance history. Most existing works primarily focus on post-training and fine-tuning tailored for cross-encoders. However, there are no post-training methods tailored for dense encoders in dialogue response selection. We argue that when the current language model, based on dense dialogue systems (such as BERT), is employed as a dense encoder, it separately encodes dialogue context and response, leading to a struggle to achieve the alignment of both representations. Thus, we propose Dial-MAE (Dialogue Contextual Masking Auto-Encoder), a straightforward yet effective post-training technique tailored for dense encoders in dialogue response selection. Dial-MAE uses an asymmetric encoder-decoder architecture to compress the dialogue semantics into dense vectors, which achieves better alignment between the features of the dialogue context and response. Our experiments have demonstrated that Dial-MAE is highly effective, achieving state-of-the-art performance on two commonly evaluated benchmarks.
comment: This paper has been accepted by NAACL 2024
♻ ☆ SoftTiger: A Clinical Foundation Model for Healthcare Workflows
We introduce SoftTiger, a clinical large language model (CLaM) designed as a foundation model for healthcare workflows. The narrative and unstructured nature of clinical notes is a major obstacle for healthcare intelligentization. We address a critical problem of structuring clinical notes into clinical data, according to international interoperability standards. We collect and annotate data for three subtasks, namely, international patient summary, clinical impression and medical encounter. We then supervised fine-tuned a state-of-the-art LLM using public and credentialed clinical data. The training is orchestrated in a way that the target model can first support basic clinical tasks such as abbreviation expansion and temporal information extraction, and then learn to perform more complex downstream clinical tasks. Moreover, we address several modeling challenges in the healthcare context, e.g., extra long context window. Our blind pairwise evaluation shows that SoftTiger outperforms other popular open-source models and GPT-3.5, comparable to Gemini-pro, with a mild gap from GPT-4. We believe that LLMs may become a step-stone towards healthcare digitalization and democratization. Therefore, we publicly release SoftTiger models at scales of 13 billion and 70 billion parameters, as well as datasets and code for our innovative scalable evaluation, hopefully, making a significant contribution to the healthcare industry.
♻ ☆ Probing Multimodal Large Language Models for Global and Local Semantic Representations LREC
The advancement of Multimodal Large Language Models (MLLMs) has greatly accelerated the development of applications in understanding integrated texts and images. Recent works leverage image-caption datasets to train MLLMs, achieving state-of-the-art performance on image-to-text tasks. However, there are few studies exploring which layers of MLLMs make the most effort to the global image information, which plays vital roles in multimodal comprehension and generation. In this study, we find that the intermediate layers of models can encode more global semantic information, whose representation vectors perform better on visual-language entailment tasks, rather than the topmost layers. We further probe models regarding local semantic representations through object recognition tasks. We find that the topmost layers may excessively focus on local information, leading to a diminished ability to encode global information. Our code and data are released via https://github.com/kobayashikanna01/probing_MLLM_rep.
comment: Accepted by LREC-COLING 2024 as a short paper (Camera Ready)
♻ ☆ In-Distribution and Out-of-Distribution Self-supervised ECG Representation Learning for Arrhythmia Detection
This paper presents a systematic investigation into the effectiveness of Self-Supervised Learning (SSL) methods for Electrocardiogram (ECG) arrhythmia detection. We begin by conducting a novel analysis of the data distributions on three popular ECG-based arrhythmia datasets: PTB-XL, Chapman, and Ribeiro. To the best of our knowledge, our study is the first to quantitatively explore and characterize these distributions in the area. We then perform a comprehensive set of experiments using different augmentations and parameters to evaluate the effectiveness of various SSL methods, namely SimCRL, BYOL, and SwAV, for ECG representation learning, where we observe the best performance achieved by SwAV. Furthermore, our analysis shows that SSL methods achieve highly competitive results to those achieved by supervised state-of-the-art methods. To further assess the performance of these methods on both In-Distribution (ID) and Out-of-Distribution (OOD) ECG data, we conduct cross-dataset training and testing experiments. Our comprehensive experiments show almost identical results when comparing ID and OOD schemes, indicating that SSL techniques can learn highly effective representations that generalize well across different OOD datasets. This finding can have major implications for ECG-based arrhythmia detection. Lastly, to further analyze our results, we perform detailed per-disease studies on the performance of the SSL methods on the three datasets.
comment: This paper has been published in the IEEE Journal of Biomedical and Health Informatics (JBHI). Copyright IEEE. Please cite as: S. Soltanieh, J. Hashemi and A. Etemad, "In-Distribution and Out-of-Distribution Self-Supervised ECG Representation Learning for Arrhythmia Detection," in IEEE Journal of Biomedical and Health Informatics, vol. 28, no. 2, pp. 789-800, Feb. 2024
♻ ☆ Imitating Cost-Constrained Behaviors in Reinforcement Learning ICAPS-24
Complex planning and scheduling problems have long been solved using various optimization or heuristic approaches. In recent years, imitation learning that aims to learn from expert demonstrations has been proposed as a viable alternative to solving these problems. Generally speaking, imitation learning is designed to learn either the reward (or preference) model or directly the behavioral policy by observing the behavior of an expert. Existing work in imitation learning and inverse reinforcement learning has focused on imitation primarily in unconstrained settings (e.g., no limit on fuel consumed by the vehicle). However, in many real-world domains, the behavior of an expert is governed not only by reward (or preference) but also by constraints. For instance, decisions on self-driving delivery vehicles are dependent not only on the route preferences/rewards (depending on past demand data) but also on the fuel in the vehicle and the time available. In such problems, imitation learning is challenging as decisions are not only dictated by the reward model but are also dependent on a cost-constrained model. In this paper, we provide multiple methods that match expert distributions in the presence of trajectory cost constraints through (a) Lagrangian-based method; (b) Meta-gradients to find a good trade-off between expected return and minimizing constraint violation; and (c) Cost-violation-based alternating gradient. We empirically show that leading imitation learning approaches imitate cost-constrained behaviors poorly and our meta-gradient-based approach achieves the best performance.
comment: Accepted to the 34th International Conference on Automated Planning and Scheduling (ICAPS-24)
♻ ☆ The Innovation Paradox: Concept Space Expansion with Diminishing Originality and the Promise of Creative AI
Innovation, typically spurred by reusing, recombining, and synthesizing existing concepts, is expected to result in an exponential growth of the concept space over time. However, our statistical analysis of TechNet, which is a comprehensive technology semantic network encompassing over four million concepts derived from patent texts, reveals a linear rather than exponential expansion of the overall technological concept space. Moreover, there is a notable decline in the originality of newly created concepts. These trends can be attributed to the constraints of human cognitive abilities to innovate beyond an ever-growing space of prior art, among other factors. Integrating creative artificial intelligence (CAI) into the innovation process holds the potential to overcome these limitations and alter the observed trends in the future.
comment: Forthcoming on the Design Science
♻ ☆ LLMs in Political Science: Heralding a New Era of Visual Analysis
Interest is increasing among political scientists in leveraging the extensive information available in images. However, the challenge of interpreting these images lies in the need for specialized knowledge in computer vision and access to specialized hardware. As a result, image analysis has been limited to a relatively small group within the political science community. This landscape could potentially change thanks to the rise of large language models (LLMs). This paper aims to raise awareness of the feasibility of using Gemini for image content analysis. A retrospective analysis was conducted on a corpus of 688 images. Content reports were elicited from Gemini for each image and then manually evaluated by the authors. We find that Gemini is highly accurate in performing object detection, which is arguably the most common and fundamental task in image analysis for political scientists. Equally important, we show that it is easy to implement as the entire command consists of a single prompt in natural language; it is fast to run and should meet the time budget of most researchers; and it is free to use and does not require any specialized hardware. In addition, we illustrate how political scientists can leverage Gemini for other image understanding tasks, including face identification, sentiment analysis, and caption generation. Our findings suggest that Gemini and other similar LLMs have the potential to drastically stimulate and accelerate image research in political science and social sciences more broadly.
comment: 7 pages, 3 tables
♻ ☆ Coarse-Tuning for Ad-hoc Document Retrieval Using Pre-trained Language Models LREC
Fine-tuning in information retrieval systems using pre-trained language models (PLM-based IR) requires learning query representations and query-document relations, in addition to downstream task-specific learning. This study introduces coarse-tuning as an intermediate learning stage that bridges pre-training and fine-tuning. By learning query representations and query-document relations in coarse-tuning, we aim to reduce the load of fine-tuning and improve the learning effect of downstream IR tasks. We propose Query-Document Pair Prediction (QDPP) for coarse-tuning, which predicts the appropriateness of query-document pairs. Evaluation experiments show that the proposed method significantly improves MRR and/or nDCG@5 in four ad-hoc document retrieval datasets. Furthermore, the results of the query prediction task suggested that coarse-tuning facilitated learning of query representation and query-document relations.
comment: Accepted at LREC-COLING 2024
♻ ☆ Look Before You Leap: Problem Elaboration Prompting Improves Mathematical Reasoning in Large Language Models
Large language models (LLMs) still grapple with complex tasks like mathematical reasoning. Despite significant efforts invested in improving prefix prompts or reasoning process, the crucial role of problem context might have been neglected. Accurate recognition of inputs is fundamental for solving mathematical tasks, as ill-formed problems could potentially mislead LLM's reasoning. In this study, we propose a new approach named Problem Elaboration Prompting (PEP) to enhance the mathematical capacities of LLMs. Specifically, PEP decomposes and elucidates the problem context before reasoning, therefore enhancing the context modeling and parsing efficiency. Experiments across datasets and models demonstrate promising performances: (1) PEP demonstrates an overall enhancement in various mathematical tasks. For instance, with the GPT-3.5 model, PEP exhibits improvements of 9.93% and 8.80% on GSM8k through greedy decoding and self-consistency, respectively. (2) PEP can be easily implemented and integrated with other prompting methods. (3) PEP shows particular strength in handling distraction problems.
♻ ☆ Follower Agnostic Methods for Stackelberg Games
In this paper, we present an efficient algorithm to solve online Stackelberg games, featuring multiple followers, in a follower-agnostic manner. Unlike previous works, our approach works even when leader has no knowledge about the followers' utility functions or strategy space. Our algorithm introduces a unique gradient estimator, leveraging specially designed strategies to probe followers. In a departure from traditional assumptions of optimal play, we model followers' responses using a convergent adaptation rule, allowing for realistic and dynamic interactions. The leader constructs the gradient estimator solely based on observations of followers' actions. We provide both non-asymptotic convergence rates to stationary points of the leader's objective and demonstrate asymptotic convergence to a \emph{local Stackelberg equilibrium}. To validate the effectiveness of our algorithm, we use this algorithm to solve the problem of incentive design on a large-scale transportation network, showcasing its robustness even when the leader lacks access to followers' demand.
comment: 31 pages
♻ ☆ Learning to Act without Actions ICLR 2024
Pre-training large models on vast amounts of web data has proven to be an effective approach for obtaining powerful, general models in domains such as language and vision. However, this paradigm has not yet taken hold in reinforcement learning. This is because videos, the most abundant form of embodied behavioral data on the web, lack the action labels required by existing methods for imitating behavior from demonstrations. We introduce Latent Action Policies (LAPO), a method for recovering latent action information, and thereby latent-action policies, world models, and inverse dynamics models, purely from videos. LAPO is the first method able to recover the structure of the true action space just from observed dynamics, even in challenging procedurally-generated environments. LAPO enables training latent-action policies that can be rapidly fine-tuned into expert-level policies, either offline using a small action-labeled dataset, or online with rewards. LAPO takes a first step towards pre-training powerful, generalist policies and world models on the vast amounts of videos readily available on the web.
comment: Accepted at ICLR 2024 (spotlight). The code can be found at http://github.com/schmidtdominik/LAPO
Optimization and Control 40
☆ The best approximation pair problem relative to two subsets in a normed space
In the classical best approximation pair (BAP) problem, one is given two nonempty, closed, convex and disjoint subsets in a finite- or an infinite-dimensional Hilbert space, and the goal is to find a pair of points, each from each subset, which realizes the distance between the subsets. This problem, which has a long history, has found applications in science and technology. We discuss the problem in more general normed spaces and with possibly non-convex subsets, and focus our attention on the issues of uniqueness and existence of the solution to the problem. To the best of our knowledge these fundamental issues have not received much attention. In particular, we present several sufficient geometric conditions for the (at most) uniqueness of a BAP relative to these subsets. These conditions are related to the structure of the boundaries of the subsets, their relative orientation, and the structure of the unit sphere of the space. In addition, we present many sufficient conditions for the existence of a BAP, possibly without convexity . Our results allow us to significantly extend the horizon of the recent alternating simultaneous Halpern-Lions-Wittmann-Bauschke (A-S-HLWB) algorithm [Censor, Mansour and Reem, The alternating simultaneous Halpern-Lions-Wittmann-Bauschke algorithm for finding the best approximation pair for two disjoint intersections of convex sets, arXiv:2304.09600 (2023)] for solving the BAP problem.
☆ A nonsmooth Frank-Wolfe algorithm through a dual cutting-plane approach
An extension of the Frank-Wolfe Algorithm (FWA), also known as Conditional Gradient algorithm, is proposed. In its standard form, the FWA allows to solve constrained optimization problems involving $\beta$-smooth cost functions, calling at each iteration a Linear Minimization Oracle. More specifically, the oracle solves a problem obtained by linearization of the original cost function. The algorithm designed and investigated in this article, named Dualized Level-Set (DLS) algorithm, extends the FWA and allows to address a class of nonsmooth costs, involving in particular support functions. The key idea behind the construction of the DLS method is a general interpretation of the FWA as a cutting-plane algorithm, from the dual point of view. The DLS algorithm essentially results from a dualization of a specific cutting-plane algorithm, based on projections on some level sets. The DLS algorithm generates a sequence of primal-dual candidates, and we prove that the corresponding primal-dual gap converges with a rate of $O(1/\sqrt{t})$.
☆ Connections between Reachability and Time Optimality
This paper presents the concept of an equivalence relation between the set of optimal control problems. By leveraging this concept, we show that the boundary of the reachability set can be constructed by the solutions of time optimal problems. Alongside, a more generalized equivalence theorem is presented together. The findings facilitate the use of solution structures from a certain class of optimal control problems to address problems in corresponding equivalent classes. As a byproduct, we state and prove the construction methods of the reachability sets of three-dimensional curves with prescribed curvature bound. The findings are twofold: Firstly, we prove that any boundary point of the reachability set, with the terminal direction taken into account, can be accessed via curves of H, CSC, CCC, or their respective subsegments, where H denotes a helicoidal arc, C a circular arc with maximum curvature, and S a straight segment. Secondly, we show that any boundary point of the reachability set, without considering the terminal direction, can be accessed by curves of CC, CS, or their respective subsegments. These findings extend the developments presented in literature regarding planar curves, or Dubins car dynamics, into spatial curves in $\mathbb{R}^3$. For higher dimensions, we confirm that the problem of identifying the reachability set of curvature bounded paths subsumes the well-known Markov-Dubins problem. These advancements in understanding the reachability of curvature bounded paths in $\mathbb{R}^3$ hold significant practical implications, particularly in the contexts of mission planning problems and time optimal guidance.
comment: Submitted to Automatica
☆ Conditional Wasserstein Distances with Applications in Bayesian OT Flow Matching
In inverse problems, many conditional generative models approximate the posterior measure by minimizing a distance between the joint measure and its learned approximation. While this approach also controls the distance between the posterior measures in the case of the Kullback--Leibler divergence, this is in general not hold true for the Wasserstein distance. In this paper, we introduce a conditional Wasserstein distance via a set of restricted couplings that equals the expected Wasserstein distance of the posteriors. Interestingly, the dual formulation of the conditional Wasserstein-1 flow resembles losses in the conditional Wasserstein GAN literature in a quite natural way. We derive theoretical properties of the conditional Wasserstein distance, characterize the corresponding geodesics and velocity fields as well as the flow ODEs. Subsequently, we propose to approximate the velocity fields by relaxing the conditional Wasserstein distance. Based on this, we propose an extension of OT Flow Matching for solving Bayesian inverse problems and demonstrate its numerical advantages on an inverse problem and class-conditional image generation.
comment: This paper supersedes arXiv:2310.13433
☆ Accelerating preconditioned ADMM via degenerate proximal point mappings
In this paper, we aim to accelerate a preconditioned alternating direction method of multipliers (pADMM), whose proximal terms are convex quadratic functions, for solving linearly constrained convex optimization problems. To achieve this, we first reformulate the pADMM into a form of proximal point method (PPM) with a positive semidefinite preconditioner which can be degenerate due to the lack of strong convexity of the proximal terms in the pADMM. Then we accelerate the pADMM by accelerating the reformulated degenerate PPM (dPPM). Specifically, we first propose an accelerated dPPM by integrating the Halpern iteration and the fast Krasnosel'ski\u{i}-Mann iteration into it, achieving asymptotic $o(1/k)$ and non-asymptotic $O(1/k)$ convergence rates. Subsequently, building upon the accelerated dPPM, we develop an accelerated pADMM algorithm that exhibits both asymptotic $o(1/k)$ and non-asymptotic $O(1/k)$ nonergodic convergence rates concerning the Karush-Kuhn-Tucker residual and the primal objective function value gap. Preliminary numerical experiments validate the theoretical findings, demonstrating that the accelerated pADMM outperforms the pADMM in solving convex quadratic programming problems.
☆ Bootstrapping Guarantees: Stability and Performance Analysis for Dynamic Encrypted Control
Encrypted dynamic controllers that operate for an unlimited time have been a challenging subject of research. The fundamental difficulty is the accumulation of errors and scaling factors in the internal state during operation. Bootstrapping, a technique commonly employed in fully homomorphic cryptosystems, can be used to avoid overflows in the controller state but can potentially introduce significant numerical errors. In this paper, we analyze dynamic encrypted control with explicit consideration of bootstrapping. By recognizing the bootstrapping errors occurring in the controller's state as an uncertainty in the robust control framework, we can provide stability and performance guarantees for the whole encrypted control system. Further, the conservatism of the stability and performance test is reduced by using a lifted version of the control system.
☆ Stability Properties of the Impulsive Goodwin's Oscillator in 1-cycle
The Impulsive Goodwin's Oscillator (IGO) is a mathematical model of a hybrid closed-loop system. It arises by closing a special kind of continuous linear positive time-invariant system with impulsive feedback, which employs both amplitude and frequency pulse modulation. The structure of IGO precludes the existence of equilibria, and all its solutions are oscillatory. With its origin in mathematical biology, the IGO also presents a control paradigm useful in a wide range of applications, in particular dosing of chemicals and medicines. Since the pulse modulation feedback mechanism introduces significant nonlinearity and non-smoothness in the closedloop dynamics, conventional controller design methods fail to apply. However, the hybrid dynamics of IGO reduce to a nonlinear, time-invariant discrete-time system, exhibiting a one-to-one correspondence between periodic solutions of the original IGO and those of the discrete-time system. The paper proposes a design approach that leverages the linearization of the equivalent discrete-time dynamics in the vicinity of a fixed point. A simple and efficient local stability condition of the 1-cycle in terms of the characteristics of the amplitude and frequency modulation functions is obtained.
comment: submitted to IEEE CDC 2024
☆ Numerical optimisation of Dirac eigenvalues
Motivated by relativistic materials, we develop a numerical scheme to support existing or state new conjectures in the spectral optimisation of eigenvalues of the Dirac operator, subject to infinite-mass boundary conditions. We study the optimality of the regular polygon (respectively, disk) among all polygons of a given number of sides (respectively, arbitrary sets), subject to area or perimeter constraints. We consider the three lowest positive eigenvalues and their ratios. Roughly, we find results analogous to known or expected for the Dirichlet Laplacian, except for the third eigenvalue which does not need to be minimised by the regular polygon (respectively, the disk) for all masses. In addition to the numerical results, a new, mass-dependent upper bound to the lowest eigenvalue in rectangles is proved and its extension to arbitrary quadrilaterals is conjectured.
comment: 19 pages, 26 figures
☆ Generalized convergence of the deep BSDE method: a step towards fully-coupled FBSDEs and applications in stochastic control
We are concerned with high-dimensional coupled FBSDE systems approximated by the deep BSDE method of Han et al. (2018). It was shown by Han and Long (2020) that the errors induced by the deep BSDE method admit a posteriori estimate depending on the loss function, whenever the backward equation only couples into the forward diffusion through the Y process. We generalize this result to fully-coupled drift coefficients, and give sufficient conditions for convergence under standard assumptions. The resulting conditions are directly verifiable for any equation. Consequently, unlike in earlier theory, our convergence analysis enables the treatment of FBSDEs stemming from stochastic optimal control problems. In particular, we provide a theoretical justification for the non-convergence of the deep BSDE method observed in recent literature, and present direct guidelines for when convergence can be guaranteed in practice. Our theoretical findings are supported by several numerical experiments in high-dimensional settings.
comment: 25 pages, 3 figures, 1 table
☆ Limited Attention Allocation in a Stochastic Linear Quadratic System with Multiplicative Noise
This study addresses limited attention allocation in a stochastic linear quadratic system with multiplicative noise. Our approach enables strategic resource allocation to enhance noise estimation and improve control decisions. We provide analytical optimal control and propose a numerical method for optimal attention allocation. Additionally, we apply our ffndings to dynamic mean-variance portfolio selection, showing effective resource allocation across time periods and factors, providing valuable insights for investors.
☆ Wirtinger gradient descent methods for low-dose Poisson phase retrieval
The problem of phase retrieval has many applications in the field of optical imaging. Motivated by imaging experiments with biological specimens, we primarily consider the setting of low-dose illumination where Poisson noise plays the dominant role. In this paper, we discuss gradient descent algorithms based on different loss functions adapted to data affected by Poisson noise, in particular in the low-dose regime. Starting from the maximum log-likelihood function for the Poisson distribution, we investigate different regularizations and approximations of the problem to design an algorithm that meets the requirements that are faced in applications. In the course of this, we focus on low-count measurements. For all suggested loss functions, we study the convergence of the respective gradient descent algorithms to stationary points and find constant step sizes that guarantee descent of the loss in each iteration. Numerical experiments in the low-dose regime are performed to corroborate the theoretical observations.
☆ Efficient Algorithms for Regularized Nonnegative Scale-invariant Low-rank Approximation Models
Regularized nonnegative low-rank approximations such as sparse Nonnegative Matrix Factorization or sparse Nonnegative Tucker Decomposition are an important branch of dimensionality reduction models with enhanced interpretability. However, from a practical perspective, the choice of regularizers and regularization coefficients, as well as the design of efficient algorithms, is challenging because of the multifactor nature of these models and the lack of theory to back these choices. This paper aims at improving upon these issues. By studying a more general model called the Homogeneous Regularized Scale-Invariant, we prove that the scale-invariance inherent to low-rank approximation models causes an implicit regularization with both unexpected beneficial and detrimental effects. This observation allows to better understand the effect of regularization functions in low-rank approximation models, to guide the choice of the regularization hyperparameters, and to design balancing strategies to enhance the convergence speed of dedicated optimization algorithms. Some of these results were already known but restricted to specific instances of regularized low-rank approximations. We also derive a generic Majorization Minimization algorithm that handles many regularized nonnegative low-rank approximations, with convergence guarantees. We showcase our contributions on sparse Nonnegative Matrix Factorization, ridge-regularized Canonical Polyadic decomposition and sparse Nonnegative Tucker Decomposition.
☆ Reweighted Quasi Norm Regularized Low-Rank Factorization for Matrix Robust PCA
Robust Principal Component Analysis (RPCA) and its associated non-convex relaxation methods constitute a significant component of matrix completion problems, wherein matrix factorization strategies effectively reduce dimensionality and enhance computational speed. However, some non-convex factorization forms lack theoretical guarantees. This paper proposes a novel strategy in non-convex quasi-norm representation, introducing a method to obtain weighted matrix quasi-norm factorization forms. Especially, explicit bilinear factor matrix factorization formulations for the weighted logarithmic norm and weighted Schatten-$q$ quasi norms with $q=1, 1/2, 2/3$ are provided, along with the establishment of corresponding matrix completion models. An Alternating Direction Method of Multipliers (ADMM) framework algorithm is employed for solving, and convergence results of the algorithm are presented.
☆ Adaptive Economic Model Predictive Control for linear systems with performance guarantees
We present a model predictive control (MPC) formulation to directly optimize economic criteria for linear constrained systems subject to disturbances and uncertain model parameters. The proposed formulation combines a certainty equivalent economic MPC with a simple least-squares parameter adaptation. For the resulting adaptive economic MPC scheme, we derive strong asymptotic and transient performance guarantees. We provide a numerical example involving building temperature control and demonstrate performance benefits of online parameter adaptation.
comment: 8 pages, 3 figures, submitted to IEEE CDC 2024
☆ Distributed Feedback Optimization of Linear Multi-agent Systems
Feedback optimization is an increasingly popular control paradigm to optimize dynamical systems, accounting for control objectives that concern the system's operation at steady-state. Existing feedback optimization techniques heavily rely on centralized system and controller architectures, and thus suffer from scalability and privacy issues when systems become large-scale. In this paper, we propose and study a distributed architecture for feedback optimization, in which each agent updates its local control state by combining the average of its neighbors with a local negative-gradient step. Under convexity and smoothness assumptions, we establish convergence of the control method to a fixed point. By reinforcing the assumptions to restricted strong convexity of the cost, we show that our algorithm converges linearly to a neighborhood of the optimal point, where the size of the neighborhood depends on the choice of the stepsize. Simulations corroborate the theoretical results.
comment: 8 pages, 4 figures
☆ The Mercer-Young Theorem for Matrix-Valued Kernels on Separable Metric Spaces
We generalize the characterization theorem going back to Mercer and Young, which states that a symmetric and continuous kernel is positive definite if and only if it is integrally positive definite. More precisely, we extend the result from real-valued kernels on compact intervals to matrix-valued kernels on separable metric spaces. We also demonstrate the applications of the generalized theorem to the field of convex optimization.
comment: 12 pages, 2 figures
☆ Computing safe bicycle routes -- Berechnung sicherer Fahrradwege
The safety of streets is difficult to quantify numerically. However, it is possible to sort streets regarding their safety into ordered categories, like safe, neutral and unsafe. In this paper we model the computation of safe bicycle routes as an optimization problem with ordinal coefficients. We describe an appropriate optimality concept for ordinal optimization problems and introduce a solution strategy for ordinal routing problems. Furthermore, we introduce a concept to incorporate safety preferences by introducing weights such that longer path with a higher safety rating are preferred. We apply the concept of ordinal routing to compute safe bicycle routes in Stuttgart, Germany, based on dates from OpenStreetMaps. We show that the choice of the weights does not only represent the trade-off of safety vs. path length, but has also an impact on the number of alternative solutions and thus on the computation time. -- Die Sicherheit von Wegen ist nur eingeschr\"ankt messbar und daher schwierig zu quantifizieren. Dahingegen ist es verh\"altnism\"a{\ss}ig leicht Wege bez\"uglich ihrer Sicherheit in geordnete Kategorien, wie beispielsweise sicher, neutral und gef\"ahrlich einzuordnen. In diesem Beitrag werden Optimierungsprobleme mit geordneten Kategorien formuliert und Optimalit\"at f\"ur diese definiert. Daraus wird eine L\"osungsstrategie f\"ur solche Probleme abgeleitet. Dar\"uber hinaus wird erkl\"art, wie die Abgrenzung zwischen den Kategorien erh\"oht werden kann, sodass l\"angere aber daf\"ur sicherere Wege mit Hilfe von Gewichten berechnet werden k\"onnen. Diese theoretischen Ergebnisse werden in der Praxis angewendet und es werden auf Grundlage von Daten von OpenStreetMaps sichere Fahrradwege in Stuttgart berechnet. Dabei zeigt sich, dass eine gute Wahl der Gewichte zu weniger L\"osungen und k\"urzeren Rechenzeiten f\"uhrt.
comment: 13 pages in German
☆ Online Prediction for Streaming Tensor Time Series
Real-time prediction plays a vital role in various control systems, such as traffic congestion control and wireless channel resource allocation. In these scenarios, the predictor usually needs to track the evolution of the latent statistical patterns in the modern high-dimensional streaming time series continuously and quickly, which presents new challenges for traditional prediction methods. This paper proposes a novel algorithm based on tensor factorization to predict streaming tensor time series online. The proposed algorithm updates the predictor in a low-complexity online manner to adapt to the time-evolving data. Additionally, an automatically adaptive version of the algorithm is presented to mitigate the negative impact of stale data. Simulation results demonstrate that our proposed methods achieve prediction accuracy similar to that of conventional offline tensor prediction methods, while being much faster than them during long-term online prediction. Therefore, our proposed algorithm provides an effective and efficient solution for the online prediction of streaming tensor time series.
☆ A Mean Field Game of Sequential Testing
We introduce a mean field game for a family of filtering problems related to the classic sequential testing of the drift of a Brownian motion. To the best of our knowledge this work presents the first treatment of mean field filtering games with stopping and an unobserved common noise in the literature. We show that the game is well-posed, characterize the solution, and establish the existence of an equilibrium under certain assumptions. We also perform numerical studies for several examples of interest.
comment: 51 pages, 3 figures
☆ Stability and convergence of the penalty formulation for nonlinear magnetostatics
The magnetostatic field distribution in a nonlinear medium amounts to the unique minimizer of the magnetic coenergy over all fields that can be generated by the same current. This is a nonlinear saddlepoint problem whose numerical solution can in principle be achieved by mixed finite element methods and appropriate nonlinear solvers. The saddlepoint structure, however, makes the solution cumbersome. A remedy is to split the magnetic field into a known source field and the gradient of a scalar potential which is governed by a convex minimization problem. The penalty approach avoids the use of artificial potentials and Lagrange multipliers and leads to an unconstrained convex minimization problem involving a large parameter. We provide a rigorous justification of the penalty approach by deriving error estimates for the approximation due to penalization. We further highlight the close connections to the Lagrange-multiplier and scalar potential approach. The theoretical results are illustrated by numerical tests for a typical benchmark problem
☆ A new dual spectral projected gradient method for log-determinant semidefinite programming with hidden clustering structures
In this paper, we propose a new efficient method for a sparse Gaussian graphical model with hidden clustering structures by extending a dual spectral projected gradient (DSPG) method proposed by Nakagaki et al.~(2020). We establish the global convergence of the proposed method to an optimal solution, and we show that the projection onto the feasible region can be solved with a low computational complexity by the use of the pool-adjacent-violators algorithm. Numerical experiments on synthesis data and real data demonstrate the efficiency of the proposed method. The proposed method takes 0.91 seconds to achieve a similar solution to the direct application of the DSPG method which takes 4361 seconds.
comment: 21 pages, 3 figures
☆ An Execution-time-certified QP Algorithm for $\ell_1$ penalty-based Soft-constrained MPC
Providing an execution time certificate and handling possible infeasibility in closed-loop are two pressing requirements of Model Predictive Control (MPC). To simultaneously meet these two requirements, this paper uses $\ell_1$ penalty-based soft-constrained MPC formulation and innovatively transforms the resulting non-smooth QP into a box-constrained QP, which is solved by our previously proposed direct and execution-time certified algorithm with only dimension-dependent (data-independent) and exact number of iterations [1]. This approach not only overcomes the limitation of our previously proposed algorithm [1], only applicable to input-constrained MPC, but also enjoys exact recovery feature (exactly recover the same solution when the original problem is feasible) of $\ell_1$ penalty-based soft-constrained MPC formulation without suffering numerical difficulty of the resulting non-smoothness. Other various real-time QP applications, not limited to MPC, will also benefit from our QP algorithm with execution-time certificate and global feasibility.
comment: 6 pages
☆ Long-Term Mine Planning with Large Neighbourhood Search
We present a Large Neighbourhood Search based approach for solving complex long-term open-pit mine planning problems. An initial feasible solution, generated by a sliding windows heuristic, is improved through repeated solves of a restricted mixed integer program. Each iteration leaves only a subset of the variables in our planning model free to take on new values. We form these subsets through the use of a novel path-based neighbourhood structure, and neighbourhood formation strategies that exploit the structure of the planning model. We show that our method is able to find near-optimal solutions to problems that cannot be solved by an off-the-shelf solver in a reasonable time, or with reasonable computational resources.
☆ Topology Optimization for the Full-Cell Design of Porous Electrodes in Electrochemical Energy Storage Devices
In this paper, we introduce a density-based topology optimization framework to design porous electrodes for maximum energy storage. We simulate the full cell with a model that incorporates electronic potential, ionic potential, and electrolyte concentration. The system consists of three materials, namely pure liquid electrolyte and the porous solids of the anode and cathode, for which we determine the optimal placement. We use separate electronic potentials to model each electrode, which allow interdigitated designs. As the result, a penalization is required to ensure that the anode and cathode do not touch, i.e. causing a short circuit. We compare multiple 2D designs generated for different fixed conditions, e.g. material properties. A 3D design with complex channel and interlocking structure is also created. All optimized designs are far superior to the traditional monolithic electrode design with respect to energy storage metrics. We observe up to 750% increase in energy storage for cases with slow effective ionic diffusion within the porous electrode.
☆ Mistake, Manipulation and Margin Guarantees in Online Strategic Classification
We consider an online strategic classification problem where each arriving agent can manipulate their true feature vector to obtain a positive predicted label, while incurring a cost that depends on the amount of manipulation. The learner seeks to predict the agent's true label given access to only the manipulated features. After the learner releases their prediction, the agent's true label is revealed. Previous algorithms such as the strategic perceptron guarantee finitely many mistakes under a margin assumption on agents' true feature vectors. However, these are not guaranteed to encourage agents to be truthful. Promoting truthfulness is intimately linked to obtaining adequate margin on the predictions, thus we provide two new algorithms aimed at recovering the maximum margin classifier in the presence of strategic agent behavior. We prove convergence, finite mistake and finite manipulation guarantees for a variety of agent cost structures. We also provide generalized versions of the strategic perceptron with mistake guarantees for different costs. Our numerical study on real and synthetic data demonstrates that the new algorithms outperform previous ones in terms of margin, number of manipulation and number of mistakes.
☆ Incentive-Compatible Vertiport Reservation in Advanced Air Mobility: An Auction-Based Approach
The rise of advanced air mobility (AAM) is expected to become a multibillion-dollar industry in the near future. Market-based mechanisms are touted to be an integral part of AAM operations, which comprise heterogeneous operators with private valuations. In this work, we study the problem of designing a mechanism to coordinate the movement of electric vertical take-off and landing (eVTOL) aircraft, operated by multiple operators each having heterogeneous valuations associated with their fleet, between vertiports, while enforcing the arrival, departure, and parking constraints at vertiports. Particularly, we propose an incentive-compatible and individually rational vertiport reservation mechanism that maximizes a social welfare metric, which encapsulates the objective of maximizing the overall valuations of all operators while minimizing the congestion at vertiports. Additionally, we improve the computational tractability of designing the reservation mechanism by proposing a mixed binary linear programming approach that is based on constructing network flow graph corresponding to the underlying problem.
comment: 26 pages, 2 figures, 1 table
☆ Incentive Designs for Learning Agents to Stabilize Coupled Exogenous Systems
We consider a large population of learning agents noncooperatively selecting strategies from a common set, influencing the dynamics of an exogenous system (ES) we seek to stabilize at a desired equilibrium. Our approach is to design a dynamic payoff mechanism capable of shaping the population's strategy profile, thus affecting the ES's state, by offering incentives for specific strategies within budget limits. Employing system-theoretic passivity concepts, we establish conditions under which a payoff mechanism can be systematically constructed to ensure the global asymptotic stabilization of the ES's equilibrium. In comparison to previous approaches originally studied in the context of the so-called epidemic population games, the method proposed here allows for more realistic epidemic models and other types of ES, such as predator-prey dynamics. Stabilization is established with the support of a Lyapunov function, which provides useful bounds on the transients.
comment: 8 pages, 3 figures
♻ ☆ On model predictive control with sampled-data input for output tracking with prescribed performance
We propose a model predictive control (MPC) scheme with sampled-data input which ensures output-reference tracking within prescribed error bounds for relative-degree-one systems. Hereby, we explicitly deduce bounds on the required maximal control input and sampling frequency such that the MPC scheme is both initially and recursively feasible. A key feature of the proposed approach is that neither terminal conditions nor a sufficiently-large prediction horizon are imposed, rendering the MPC scheme computationally efficient. We illustrate the MPC algorithm via a numerical example of a torsional oscillator.
♻ ☆ Nonlinear Control Allocation: A Learning Based Approach IEEE Conference on Decision and Control
Modern aircraft are designed with redundant control effectors to cater for fault tolerance and maneuverability requirements. This leads to aircraft being over-actuated and requires control allocation schemes to distribute the control commands among control effectors. Traditionally, optimization-based control allocation schemes are used; however, for nonlinear allocation problems, these methods require large computational resources. In this work, an artificial neural network (ANN) based nonlinear control allocation scheme is proposed. The proposed scheme is composed of learning the inverse of the control effectiveness map through ANN, and then implementing it as an allocator instead of solving an online optimization problem. Stability conditions are presented for closed-loop systems incorporating the allocator, and computational challenges are explored with piece-wise linear effectiveness functions and ANN-based allocators. To demonstrate the efficacy of the proposed scheme, it is compared with a standard quadratic programming-based method for control allocation.
comment: submitted to IEEE Conference on Decision and Control (CDC), 2024
♻ ☆ Stochastic Approximation with Delayed Updates: Finite-Time Rates under Markovian Sampling AISTATS
Motivated by applications in large-scale and multi-agent reinforcement learning, we study the non-asymptotic performance of stochastic approximation (SA) schemes with delayed updates under Markovian sampling. While the effect of delays has been extensively studied for optimization, the manner in which they interact with the underlying Markov process to shape the finite-time performance of SA remains poorly understood. In this context, our first main contribution is to show that under time-varying bounded delays, the delayed SA update rule guarantees exponentially fast convergence of the \emph{last iterate} to a ball around the SA operator's fixed point. Notably, our bound is \emph{tight} in its dependence on both the maximum delay $\tau_{max}$, and the mixing time $\tau_{mix}$. To achieve this tight bound, we develop a novel inductive proof technique that, unlike various existing delayed-optimization analyses, relies on establishing uniform boundedness of the iterates. As such, our proof may be of independent interest. Next, to mitigate the impact of the maximum delay on the convergence rate, we provide the first finite-time analysis of a delay-adaptive SA scheme under Markovian sampling. In particular, we show that the exponent of convergence of this scheme gets scaled down by $\tau_{avg}$, as opposed to $\tau_{max}$ for the vanilla delayed SA rule; here, $\tau_{avg}$ denotes the average delay across all iterations. Moreover, the adaptive scheme requires no prior knowledge of the delay sequence for step-size tuning. Our theoretical findings shed light on the finite-time effects of delays for a broad class of algorithms, including TD learning, Q-learning, and stochastic gradient descent under Markovian sampling.
comment: Accepted to the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024!
♻ ☆ Tikhonov Regularization for Stochastic Non-Smooth Convex Optimization in Hilbert Spaces
To solve non-smooth convex optimization problems with a noisy gradient input, we analyze the global behavior of subgradient-like flows under stochastic errors. The objective function is composite, being equal to the sum of two convex functions, one being differentiable and the other potentially non-smooth. We then use stochastic differential inclusions where the drift term is minus the subgradient of the objective function, and the diffusion term is either bounded or square-integrable. In this context, under Lipschitz's continuity of the differentiable term and a growth condition of the non-smooth term, our first main result shows almost sure weak convergence of the trajectory process towards a minimizer of the objective function. Then, using Tikhonov regularization with a properly tuned vanishing parameter, we can obtain almost sure strong convergence of the trajectory towards the minimum norm solution. We find an explicit tuning of this parameter when our objective function satisfies a local error-bound inequality. We also provide a comprehensive complexity analysis by establishing several new pointwise and ergodic convergence rates in expectation for the convex and strongly convex case.
comment: 34 pages, 2 tables. arXiv admin note: text overlap with arXiv:2207.02750
♻ ☆ Fast Convergence of Inertial Multiobjective Gradient-like Systems with Asymptotic Vanishing Damping
We present a new gradient-like dynamical system related to unconstrained convex smooth multiobjective optimization which involves inertial effects and asymptotic vanishing damping. To the best of our knowledge, this system is the first inertial gradient-like system for multiobjective optimization problems including asymptotic vanishing damping, expanding the ideas laid out in [H. Attouch and G. Garrigos, Multiobjective optimization: an inertial approach to Pareto optima, preprint, arXiv:1506.02823, 201]. We prove existence of solutions to this system in finite dimensions and further prove that its bounded solutions converge weakly to weakly Pareto optimal points. In addition, we obtain a convergence rate of order $O(t^{-2})$ for the function values measured with a merit function. This approach presents a good basis for the development of fast gradient methods for multiobjective optimization.
comment: 25 pages, 3 Figures
♻ ☆ Redesigning Large-Scale Multimodal Transit Networks with Shared Autonomous Mobility Services
This study addresses a large-scale multimodal transit network design problem, with Shared Autonomous Mobility Services (SAMS) as both transit feeders and an origin-to-destination mode. The framework captures spatial demand and modal characteristics, considers intermodal transfers and express services, determines transit infrastructure investment and path flows, and generates transit routes. A system-optimal multimodal transit network is designed with minimum total door-to-door generalized costs of users and operators, satisfying transit origin-destination demand within a pre-set infrastructure budget. Firstly, the geography, demand, and modes in each zone are characterized with continuous approximation. The decisions of network link investment and multimodal path flows in zonal connection optimization are formulated as a minimum-cost multi-commodity network flow (MCNF) problem and solved efficiently with a mixed-integer linear programming (MILP) solver. Subsequently, the route generation problem is solved by expanding the MCNF formulation to minimize intramodal transfers. The model is illustrated through a set of experiments with the Chicago network comprised of 50 zones and seven modes, under three scenarios. The computational results present savings in traveler journey time and operator cost demonstrating the potential benefits of collaboration between multimodal transit systems and SAMS.
comment: 48 pages, 18 figures, accepted for publication in Transportation Research Part C: Emerging Technologies, and presentation in the 25th International Symposium on Transportation and Traffic Theory (ISTTT25)
♻ ☆ Parabolic PDE-constrained optimal control under uncertainty with entropic risk measure using quasi-Monte Carlo integration
We study the application of a tailored quasi-Monte Carlo (QMC) method to a class of optimal control problems subject to parabolic partial differential equation (PDE) constraints under uncertainty: the state in our setting is the solution of a parabolic PDE with a random thermal diffusion coefficient, steered by a control function. To account for the presence of uncertainty in the optimal control problem, the objective function is composed with a risk measure. We focus on two risk measures, both involving high-dimensional integrals over the stochastic variables: the expected value and the (nonlinear) entropic risk measure. The high-dimensional integrals are computed numerically using specially designed QMC methods and, under moderate assumptions on the input random field, the error rate is shown to be essentially linear, independently of the stochastic dimension of the problem -- and thereby superior to ordinary Monte Carlo methods. Numerical results demonstrate the effectiveness of our method.
♻ ☆ A novel strong duality-based reformulation for trilevel infrastructure models in energy systems development
We explore the class of trilevel equilibrium problems with a focus on energy-environmental applications and present a novel single-level reformulation for such problems, based on strong duality. To the best of our knowledge, only one alternative single-level reformulation for trilevel problems exists. This reformulation uses a representation of the bottom-level solution set, whereas we propose a reformulation based on strong duality. Our novel reformulation is compared to this existing formulation, discussing both model sizes and computational performance. In particular, we apply this trilevel framework to a power market model, exploring the possibilities of an international policymaker in reducing emissions of the system. Using the proposed methods, we are able to obtain globally optimal solutions for a five-node case study representing the Nordic countries and assess the impact of a carbon tax on the electricity production portfolio.
comment: 22 pages, 4 figures
♻ ☆ Infinite horizon McKean-Vlasov FBSDEs and applications to mean field control problems
In this paper, we study a class of infinite horizon fully coupled McKean-Vlasov forward-backward stochastic differential equations (FBSDEs). We propose a generalized monotonicity condition involving two flexible functions. Under this condition, we establish the well-posedness results for infinite horizon McKean-Vlasov FBSDEs by the method of continuation, including the unique solvability, an estimate of the solution, and the related continuous dependence property of the solution on the coefficients. Based on the solvability result, we study an infinite horizon mean field control problem. Moreover, by choosing appropriate form of the flexible functions, we can eliminate the different phenomenon between the linear-quadratic (LQ) problems on infinite horizon and finite horizon proposed in Wei and Yu (SIAM J. Control Optim. 59: 2594-2623, 2021).
♻ ☆ A Mixed-Integer Linear Program to create the shifts in a supermarket
The shift design and the personnel scheduling problem is known to be a difficult problem. It is a real-world problem which has lots of applications in the organization of companies. Solutions are usually found by dividing the problem in two steps: first the shifts are created, then the employees are assigned to them by respecting a bunch of constraints. The assignment of different tasks increases the complexity, since we have to consider the skills of the single employee necessary to perform any activity. In this paper we present a mixed-integer linear programming formulation which models together the shift creation and the construction of rosters for employees, with the objective of minimizing the amount of uncovered demand. Finally we provide the results for three real-world instances, confirming that this approach is promising.
♻ ☆ Sample Efficient Reinforcement Learning with Partial Dynamics Knowledge AAAI
The problem of sample complexity of online reinforcement learning is often studied in the literature without taking into account any partial knowledge about the system dynamics that could potentially accelerate the learning process. In this paper, we study the sample complexity of online Q-learning methods when some prior knowledge about the dynamics is available or can be learned efficiently. We focus on systems that evolve according to an additive disturbance model of the form $S_{h+1} = f(S_h, A_h) + W_h$, where $f$ represents the underlying system dynamics, and $W_h$ are unknown disturbances independent of states and actions. In the setting of finite episodic Markov decision processes with $S$ states, $A$ actions, and episode length $H$, we present an optimistic Q-learning algorithm that achieves $\tilde{\mathcal{O}}(\text{Poly}(H)\sqrt{T})$ regret under perfect knowledge of $f$, where $T$ is the total number of interactions with the system. This is in contrast to the typical $\tilde{\mathcal{O}}(\text{Poly}(H)\sqrt{SAT})$ regret for existing Q-learning methods. Further, if only a noisy estimate $\hat{f}$ of $f$ is available, our method can learn an approximately optimal policy in a number of samples that is independent of the cardinalities of state and action spaces. The sub-optimality gap depends on the approximation error $\hat{f}-f$, as well as the Lipschitz constant of the corresponding optimal value function. Our approach does not require modeling of the transition probabilities and enjoys the same memory complexity as model-free methods.
comment: Published in the 38th Annual AAAI Conference on Artificial Intelligence
♻ ☆ Follower Agnostic Methods for Stackelberg Games
In this paper, we present an efficient algorithm to solve online Stackelberg games, featuring multiple followers, in a follower-agnostic manner. Unlike previous works, our approach works even when leader has no knowledge about the followers' utility functions or strategy space. Our algorithm introduces a unique gradient estimator, leveraging specially designed strategies to probe followers. In a departure from traditional assumptions of optimal play, we model followers' responses using a convergent adaptation rule, allowing for realistic and dynamic interactions. The leader constructs the gradient estimator solely based on observations of followers' actions. We provide both non-asymptotic convergence rates to stationary points of the leader's objective and demonstrate asymptotic convergence to a \emph{local Stackelberg equilibrium}. To validate the effectiveness of our algorithm, we use this algorithm to solve the problem of incentive design on a large-scale transportation network, showcasing its robustness even when the leader lacks access to followers' demand.
comment: 31 pages
♻ ☆ A Polynomial-time Algorithm for the Large Scale of Airplane Refueling Problem
Airplane refueling problem is a nonlinear unconstrained optimization problem with $n!$ feasible solutions. Given a fleet of $n$ airplanes with mid-air refueling technique, the question is to find the best refueling policy to make the last remaining airplane travels the farthest. In order to deal with the large scale of airplanes refueling instances, we proposed the definition of sequential feasible solution by employing the refueling properties of data structure. We proved that if an airplanes refueling instance has feasible solutions, it must have the sequential feasible solutions; and the optimal feasible solution must be the optimal sequential feasible solution. Then we proposed the sequential search algorithm which consists of two steps. The first step of the sequential search algorithm aims to seek out all of the sequential feasible solutions. When the input size of $n$ is greater than an index number, we proved that the number of the sequential feasible solutions will change to grow at a polynomial rate. The second step of the sequential search algorithm aims to search for the maximal sequential feasible solution by bubble sorting all of the sequential feasible solutions. Moreover, we built an efficient computability scheme, according to which we could forecast within a polynomial time the computational complexity of the sequential search algorithm that runs on any given airplanes refueling instance. Thus we could provide a computational strategy for decision makers or algorithm users by considering with their available computing resources.
comment: 18 pages, 2 figures
Systems and Control 39
☆ Usage-Specific Survival Modeling Based on Operational Data and Neural Networks
Accurate predictions of when a component will fail are crucial when planning maintenance, and by modeling the distribution of these failure times, survival models have shown to be particularly useful in this context. The presented methodology is based on conventional neural network-based survival models that are trained using data that is continuously gathered and stored at specific times, called snapshots. An important property of this type of training data is that it can contain more than one snapshot from a specific individual which results in that standard maximum likelihood training can not be directly applied since the data is not independent. However, the papers show that if the data is in a specific format where all snapshot times are the same for all individuals, called homogeneously sampled, maximum likelihood training can be applied and produce desirable results. In many cases, the data is not homogeneously sampled and in this case, it is proposed to resample the data to make it homogeneously sampled. How densely the dataset is sampled turns out to be an important parameter; it should be chosen large enough to produce good results, but this also increases the size of the dataset which makes training slow. To reduce the number of samples needed during training, the paper also proposes a technique to, instead of resampling the dataset once before the training starts, randomly resample the dataset at the start of each epoch during the training. The proposed methodology is evaluated on both a simulated dataset and an experimental dataset of starter battery failures. The results show that if the data is homogeneously sampled the methodology works as intended and produces accurate survival models. The results also show that randomly resampling the dataset on each epoch is an effective way to reduce the size of the training data.
comment: 7 pages
☆ Connections between Reachability and Time Optimality
This paper presents the concept of an equivalence relation between the set of optimal control problems. By leveraging this concept, we show that the boundary of the reachability set can be constructed by the solutions of time optimal problems. Alongside, a more generalized equivalence theorem is presented together. The findings facilitate the use of solution structures from a certain class of optimal control problems to address problems in corresponding equivalent classes. As a byproduct, we state and prove the construction methods of the reachability sets of three-dimensional curves with prescribed curvature bound. The findings are twofold: Firstly, we prove that any boundary point of the reachability set, with the terminal direction taken into account, can be accessed via curves of H, CSC, CCC, or their respective subsegments, where H denotes a helicoidal arc, C a circular arc with maximum curvature, and S a straight segment. Secondly, we show that any boundary point of the reachability set, without considering the terminal direction, can be accessed by curves of CC, CS, or their respective subsegments. These findings extend the developments presented in literature regarding planar curves, or Dubins car dynamics, into spatial curves in $\mathbb{R}^3$. For higher dimensions, we confirm that the problem of identifying the reachability set of curvature bounded paths subsumes the well-known Markov-Dubins problem. These advancements in understanding the reachability of curvature bounded paths in $\mathbb{R}^3$ hold significant practical implications, particularly in the contexts of mission planning problems and time optimal guidance.
comment: Submitted to Automatica
☆ Fpga-Based Neural Thrust Controller for UAVs
The advent of unmanned aerial vehicles (UAVs) has improved a variety of fields by providing a versatile, cost-effective and accessible platform for implementing state-of-the-art algorithms. To accomplish a broader range of tasks, there is a growing need for enhanced on-board computing to cope with increasing complexity and dynamic environmental conditions. Recent advances have seen the application of Deep Neural Networks (DNNs), particularly in combination with Reinforcement Learning (RL), to improve the adaptability and performance of UAVs, especially in unknown environments. However, the computational requirements of DNNs pose a challenge to the limited computing resources available on many UAVs. This work explores the use of Field Programmable Gate Arrays (FPGAs) as a viable solution to this challenge, offering flexibility, high performance, energy and time efficiency. We propose a novel hardware board equipped with an Artix-7 FPGA for a popular open-source micro-UAV platform. We successfully validate its functionality by implementing an RL-based low-level controller using real-world experiments.
☆ An Efficient Risk-aware Branch MPC for Automated Driving that is Robust to Uncertain Vehicle Behaviors
One of the critical challenges in automated driving is ensuring safety of automated vehicles despite the unknown behavior of the other vehicles. Although motion prediction modules are able to generate a probability distribution associated with various behavior modes, their probabilistic estimates are often inaccurate, thus leading to a possibly unsafe trajectory. To overcome this challenge, we propose a risk-aware motion planning framework that appropriately accounts for the ambiguity in the estimated probability distribution. We formulate the risk-aware motion planning problem as a min-max optimization problem and develop an efficient iterative method by incorporating a regularization term in the probability update step. Via extensive numerical studies, we validate the convergence of our method and demonstrate its advantages compared to the state-of-the-art approaches.
☆ Neural Network-Based Piecewise Survival Models
In this paper, a family of neural network-based survival models is presented. The models are specified based on piecewise definitions of the hazard function and the density function on a partitioning of the time; both constant and linear piecewise definitions are presented, resulting in a family of four models. The models can be seen as an extension of the commonly used discrete-time and piecewise exponential models and thereby add flexibility to this set of standard models. Using a simulated dataset the models are shown to perform well compared to the highly expressive, state-of-the-art energy-based model, while only requiring a fraction of the computation time.
comment: 7 pages
MPC-CBF with Adaptive Safety Margins for Safety-critical Teleoperation over Imperfect Network Connections
The paper focuses on the design of a control strategy for safety-critical remote teleoperation. The main goal is to make the controlled system track the desired velocity specified by an operator while avoiding obstacles despite communication delays. Control Barrier Functions (CBFs) are used to define the safety constraints that the system has to respect to avoid obstacles, while Model Predictive Control (MPC) provides the framework for adjusting the desired input, taking the constraints into account. The resulting input is sent to the remote system, where appropriate low-level velocity controllers translate it into system-specific commands. The main novelty of the paper is a method to make the CBFs robust against the uncertainties caused by the network delays affecting the system's state and do so in a less conservative manner. The results show how the proposed method successfully solves the safety-critical teleoperation problem, making the controlled systems avoid obstacles with different types of network delay. The controller has also been tested in simulation and on a real manipulator, demonstrating its general applicability when reliable low-level velocity controllers are available.
comment: Accepted for publication in the 2024 European Control Conference (ECC)
☆ Addressing Data Annotation Challenges in Multiple Sensors: A Solution for Scania Collected Datasets
Data annotation in autonomous vehicles is a critical step in the development of Deep Neural Network (DNN) based models or the performance evaluation of the perception system. This often takes the form of adding 3D bounding boxes on time-sequential and registered series of point-sets captured from active sensors like Light Detection and Ranging (LiDAR) and Radio Detection and Ranging (RADAR). When annotating multiple active sensors, there is a need to motion compensate and translate the points to a consistent coordinate frame and timestamp respectively. However, highly dynamic objects pose a unique challenge, as they can appear at different timestamps in each sensor's data. Without knowing the speed of the objects, their position appears to be different in different sensor outputs. Thus, even after motion compensation, highly dynamic objects are not matched from multiple sensors in the same frame, and human annotators struggle to add unique bounding boxes that capture all objects. This article focuses on addressing this challenge, primarily within the context of Scania collected datasets. The proposed solution takes a track of an annotated object as input and uses the Moving Horizon Estimation (MHE) to robustly estimate its speed. The estimated speed profile is utilized to correct the position of the annotated box and add boxes to object clusters missed by the original annotation.
comment: Accepted to European Control Conference 2024
Optimal Control Synthesis of Markov Decision Processes for Efficiency with Surveillance Tasks
We investigate the problem of optimal control synthesis for Markov Decision Processes (MDPs), addressing both qualitative and quantitative objectives. Specifically, we require the system to fulfill a qualitative surveillance task in the sense that a specific region of interest can be visited infinitely often with probability one. Furthermore, to quantify the performance of the system, we consider the concept of efficiency, which is defined as the ratio between rewards and costs. This measure is more general than the standard long-run average reward metric as it aims to maximize the reward obtained per unit cost. Our objective is to synthesize a control policy that ensures the surveillance task while maximizes the efficiency. We provide an effective approach to synthesize a stationary control policy achieving $\epsilon$-optimality by integrating state classifications of MDPs and perturbation analysis in a novel manner. Our results generalize existing works on efficiency-optimal control synthesis for MDP by incorporating qualitative surveillance tasks. A robot motion planning case study is provided to illustrate the proposed algorithm.
☆ From Virtual Reality to the Emerging Discipline of Perception Engineering
This paper makes the case that a powerful new discipline, which we term perception engineering, is steadily emerging. It follows from a progression of ideas that involve creating illusions, from historical paintings and film, to video games and virtual reality in modern times. Rather than creating physical artifacts such as bridges, airplanes, or computers, perception engineers create illusory perceptual experiences. The scope is defined over any agent that interacts with the physical world, including both biological organisms (humans, animals) and engineered systems (robots, autonomous systems). The key idea is that an agent, called a producer, alters the environment with the intent to alter the perceptual experience of another agent, called a receiver. Most importantly, the paper introduces a precise mathematical formulation of this process, based on the von Neumann-Morgenstern notion of information, to help scope and define the discipline. It is then applied to the cases of engineered and biological agents with discussion of its implications on existing fields such as virtual reality, robotics, and even social media. Finally, open challenges and opportunities for involvement are identified.
comment: 30 pages, 5 figures
☆ Bootstrapping Guarantees: Stability and Performance Analysis for Dynamic Encrypted Control
Encrypted dynamic controllers that operate for an unlimited time have been a challenging subject of research. The fundamental difficulty is the accumulation of errors and scaling factors in the internal state during operation. Bootstrapping, a technique commonly employed in fully homomorphic cryptosystems, can be used to avoid overflows in the controller state but can potentially introduce significant numerical errors. In this paper, we analyze dynamic encrypted control with explicit consideration of bootstrapping. By recognizing the bootstrapping errors occurring in the controller's state as an uncertainty in the robust control framework, we can provide stability and performance guarantees for the whole encrypted control system. Further, the conservatism of the stability and performance test is reduced by using a lifted version of the control system.
☆ Formal Verification with Constrained Polynomial Logical Zonotope
In this paper, we propose using constrained polynomial logical zonotopes for formal verification of logical systems. We perform reachability analysis to compute the set of states that could be reached. To do this, we utilize a recently introduced set representation called polynomial logical zonotopes for performing computationally efficient and exact reachability analysis on logical systems. Notably, polynomial logical zonotopes address the "curse of dimensionality" when analyzing the reachability of logical systems since the set representation can represent 2^n binary vectors using n generators. After finishing the reachability analysis, the formal verification involves verifying whether the intersection of the calculated reachable set and the unsafe set is empty or not. However, polynomial logical zonotopes are not closed under intersections. To address this, we formulate constrained polynomial logical zonotopes, which maintain the computational efficiency and exactness of polynomial logical zonotopes for reachability analysis while supporting exact intersections. Furthermore, we present an extensive empirical study illustrating and verifying the benefits of using constrained polynomial logical zonotopes for the formal verification of logical systems.
☆ A Dynamic Programming Approach for Road Traffic Estimation
We consider a road network represented by a directed graph. We assume to collect many measurements of traffic flows on all the network arcs, or on a subset of them. We assume that the users are divided into different groups. Each group follows a different path. The flows of all user groups are modeled as a set of independent Poisson processes. Our focus is estimating the paths followed by each user group, and the means of the associated Poisson processes. We present a possible solution based on a Dynamic Programming algorithm. The method relies on the knowledge of high order cumulants. We discuss the theoretical properties of the introduced method. Finally, we present some numerical tests on well-known benchmark networks, using synthetic data.
☆ Stability Properties of the Impulsive Goodwin's Oscillator in 1-cycle
The Impulsive Goodwin's Oscillator (IGO) is a mathematical model of a hybrid closed-loop system. It arises by closing a special kind of continuous linear positive time-invariant system with impulsive feedback, which employs both amplitude and frequency pulse modulation. The structure of IGO precludes the existence of equilibria, and all its solutions are oscillatory. With its origin in mathematical biology, the IGO also presents a control paradigm useful in a wide range of applications, in particular dosing of chemicals and medicines. Since the pulse modulation feedback mechanism introduces significant nonlinearity and non-smoothness in the closedloop dynamics, conventional controller design methods fail to apply. However, the hybrid dynamics of IGO reduce to a nonlinear, time-invariant discrete-time system, exhibiting a one-to-one correspondence between periodic solutions of the original IGO and those of the discrete-time system. The paper proposes a design approach that leverages the linearization of the equivalent discrete-time dynamics in the vicinity of a fixed point. A simple and efficient local stability condition of the 1-cycle in terms of the characteristics of the amplitude and frequency modulation functions is obtained.
comment: submitted to IEEE CDC 2024
☆ Safe and Robust Reinforcement-Learning: Principles and Practice
Reinforcement Learning (RL) has shown remarkable success in solving relatively complex tasks, yet the deployment of RL systems in real-world scenarios poses significant challenges related to safety and robustness. This paper aims to identify and further understand those challenges thorough the exploration of the main dimensions of the safe and robust RL landscape, encompassing algorithmic, ethical, and practical considerations. We conduct a comprehensive review of methodologies and open problems that summarizes the efforts in recent years to address the inherent risks associated with RL applications. After discussing and proposing definitions for both safe and robust RL, the paper categorizes existing research works into different algorithmic approaches that enhance the safety and robustness of RL agents. We examine techniques such as uncertainty estimation, optimisation methodologies, exploration-exploitation trade-offs, and adversarial training. Environmental factors, including sim-to-real transfer and domain adaptation, are also scrutinized to understand how RL systems can adapt to diverse and dynamic surroundings. Moreover, human involvement is an integral ingredient of the analysis, acknowledging the broad set of roles that humans can take in this context. Importantly, to aid practitioners in navigating the complexities of safe and robust RL implementation, this paper introduces a practical checklist derived from the synthesized literature. The checklist encompasses critical aspects of algorithm design, training environment considerations, and ethical guidelines. It will serve as a resource for developers and policymakers alike to ensure the responsible deployment of RL systems in many application domains.
☆ Feedback Linearizable Discretizations of Second Order Mechanical Systems using Retraction Maps
Mechanical systems, in nature, are often described by a set of continuous-time, nonlinear, second-order differential equations (SODEs). This has motivated designs of various control laws implemented on digital controllers, consequently requiring numerical discretization schemes. Feedback linearizability of such sampled systems depends on the discretization scheme or map choice. In this article, we utilize retraction maps and their lifts to construct feedback linearizable discretizations for SODEs, which can be applied to various mechanical systems.
☆ HyRRT-Connect: A Bidirectional Rapidly-Exploring Random Trees Motion Planning Algorithm for Hybrid Systems
This paper proposes a bidirectional rapidly-exploring random trees (RRT) algorithm to solve the motion planning problem for hybrid systems. The proposed algorithm, called HyRRT-Connect, propagates in both forward and backward directions in hybrid time until an overlap between the forward and backward propagation results is detected. Then, HyRRT-Connect constructs a motion plan through the reversal and concatenation of functions defined on hybrid time domains, ensuring the motion plan thoroughly satisfies the given hybrid dynamics. To address the potential discontinuity along the flow caused by tolerating some distance between the forward and backward partial motion plans, we reconstruct the backward partial motion plan by a forward-in-hybrid-time simulation from the final state of the forward partial motion plan. By applying the reversed input of the backward partial motion plan, the reconstruction process effectively eliminates the discontinuity and ensures that as the tolerance distance decreases to zero, the distance between the endpoint of the reconstructed motion plan and the final state set approaches zero. The proposed algorithm is applied to an actuated bouncing ball example and a walking robot example so as to highlight its generality and computational improvement.
comment: Accepted by the 8th IFAC International Conference on Analysis and Design of Hybrid Systems (ADHS 2024)
☆ Adaptive Economic Model Predictive Control for linear systems with performance guarantees
We present a model predictive control (MPC) formulation to directly optimize economic criteria for linear constrained systems subject to disturbances and uncertain model parameters. The proposed formulation combines a certainty equivalent economic MPC with a simple least-squares parameter adaptation. For the resulting adaptive economic MPC scheme, we derive strong asymptotic and transient performance guarantees. We provide a numerical example involving building temperature control and demonstrate performance benefits of online parameter adaptation.
comment: 8 pages, 3 figures, submitted to IEEE CDC 2024
☆ Multivariable control of modular multilevel converters with convergence and safety guarantees
Well-designed current control is a key factor in ensuring the efficient and safe operation of modular multilevel converters (MMCs). Even though this control problem involves multiple control objectives, conventional current control schemes are comprised of independently designed decoupled controllers, e.g., proportional-integral (PI) or proportional-resonant (PR). Due to the bilinearity of the MMC dynamics, tuning PI and PR controllers so that good performance and constraint satisfaction are guaranteed is quite challenging. This challenge becomes more relevant in an AC/AC MMC configuration due to the complexity of tracking the single-phase sinusoidal components of the MMC output. In this paper, we propose a method to design a multivariable controller, i.e., a static feedback gain, to regulate the MMC currents. We use a physics-informed transformation to model the MMC dynamics linearly and synthesise the proposed controller. We use this linear model to formulate a linear matrix inequality that computes a feedback gain that guarantees safe and effective operation, including (i) limited tracking error, (ii) stability, and (iii) meeting all constraints. To test the efficacy of our method, we examine its performance in a direct AC/AC MMC simulated in Simulink/PLECS and in a scaled-down AC/AC MMC prototype to investigate the ultra-fast charging of electric vehicles.
comment: Submitted to IEEE Open Journal of the Industrial Electronics
Differentially Private Dual Gradient Tracking for Distributed Resource Allocation
This paper investigates privacy issues in distributed resource allocation over directed networks, where each agent holds a private cost function and optimizes its decision subject to a global coupling constraint through local interaction with other agents. Conventional methods for resource allocation over directed networks require all agents to transmit their original data to neighbors, which poses the risk of disclosing sensitive and private information. To address this issue, we propose an algorithm called differentially private dual gradient tracking (DP-DGT) for distributed resource allocation, which obfuscates the exchanged messages using independent Laplacian noise. Our algorithm ensures that the agents' decisions converge to a neighborhood of the optimal solution almost surely. Furthermore, without the assumption of bounded gradients, we prove that the cumulative differential privacy loss under the proposed algorithm is finite even when the number of iterations goes to infinity. To the best of our knowledge, we are the first to simultaneously achieve these two goals in distributed resource allocation problems over directed networks. Finally, numerical simulations on economic dispatch problems within the IEEE 14-bus system illustrate the effectiveness of our proposed algorithm.
Differentially Private Distributed Nonconvex Stochastic Optimization with Quantized Communications
This paper proposes a new distributed nonconvex stochastic optimization algorithm that can achieve privacy protection, communication efficiency and convergence simultaneously. Specifically, each node adds time-varying privacy noises to its local state to avoid information leakage, and then quantizes its noise-perturbed state before transmitting to improve communication efficiency. By employing the subsampling method controlled through the sample-size parameter, the proposed algorithm reduces the impact of privacy noises, and enhances the differential privacy level. When the global cost function satisfies the Polyak-Lojasiewicz condition, the mean and high-probability convergence rate and the oracle complexity of the proposed algorithm are given. Importantly, the proposed algorithm achieves both the mean convergence and a finite cumulative differential privacy budget over infinite iterations as the sample-size goes to infinity. A numerical example of the distributed training on the "MNIST" dataset is given to show the effectiveness of the algorithm.
☆ Linear Hybrid Asymmetrical Load-Modulated Balanced Amplifier with Multi-Band Reconfigurability and Antenna-VSWR Resilience
This paper presents the first-ever highly linear and load-insensitive three-way load-modulation power amplifier (PA) based on reconfigurable hybrid asymmetrical load modulated balanced amplifier (H-ALMBA). Through proper amplitude and phase controls, the carrier, control amplifier (CA), and two peaking balanced amplifiers (BA1 and BA2) can form a linear high-order load modulation over wide bandwidth. Moreover, it is theoretically unveiled that the load modulation behavior of H-ALMBA can be insensitive to load mismatch by leveraging bias reconfiguration and the intrinsic load-insensitivity of balanced topology. Specifically, the PA's linearity and efficiency profiles can be maintained against arbitrary load mismatch through $Z_\mathrm{L}$-dependent reconfiguration of CA supply voltage ($V_\mathrm{DD,CA}$) and turning-on sequence of BA1 and BA2. Based on the proposed theory, an RF-input linear H-ALMBA is developed with GaN transistors and wideband quadrature hybrids. Over the design bandwidth from $1.7$-$2.9$ GHz, an efficiency of $56.8\%$$-$$72.9\%$ at peak power and $49.8\%$$-$$61.2\%$ at $10$-dB PBO are measured together with linear AMAM and AMPM responses. In modulated evaluation with 4G LTE signal, an EVM of $3.1\%$, ACPR of $-39$ dB, and average efficiency of up to $52\%$ are measured. Moreover, the reconfigurable H-ALMBA experimentally maintains an excellent average efficiency and linearity against arbitrary load mismatch at $2:1$ VSWR, and this mismatch-resilient operation can be achieved at any in-band frequencies. The overall measured performance favorably outperforms the state-of-the-art.
comment: This work has been submitted to the IEEE for possible publication
☆ An Execution-time-certified QP Algorithm for $\ell_1$ penalty-based Soft-constrained MPC
Providing an execution time certificate and handling possible infeasibility in closed-loop are two pressing requirements of Model Predictive Control (MPC). To simultaneously meet these two requirements, this paper uses $\ell_1$ penalty-based soft-constrained MPC formulation and innovatively transforms the resulting non-smooth QP into a box-constrained QP, which is solved by our previously proposed direct and execution-time certified algorithm with only dimension-dependent (data-independent) and exact number of iterations [1]. This approach not only overcomes the limitation of our previously proposed algorithm [1], only applicable to input-constrained MPC, but also enjoys exact recovery feature (exactly recover the same solution when the original problem is feasible) of $\ell_1$ penalty-based soft-constrained MPC formulation without suffering numerical difficulty of the resulting non-smoothness. Other various real-time QP applications, not limited to MPC, will also benefit from our QP algorithm with execution-time certificate and global feasibility.
comment: 6 pages
☆ Fault-tolerant properties of scale-free linear protocols for synchronization of homogeneous multi-agent systems
Originally, protocols were designed for multi-agent systems (MAS) using information about the network. However, in many cases there is no or only limited information available about the network. Recently, there has been a focus on scale-free synchronization of multi-agent systems (MAS). In this case, the protocol is designed without any prior information about the network. As long as the network contains a directed spanning tree, the scale-free protocol guarantees that the network achieves synchronization. If there is no directed spanning tree for the network then synchronization cannot be achieved. But what happens when these scale-free protocols are applied to such a network where the directed spanning tree no longer exists? The latter might arise if, for instance, a fault occurs in one of more crucial links. This paper establishes that the network decomposes into a number of basic bicomponents which achieves synchronization among all nodes in this basic bicomponent. On the other hand, nodes which are not part of any basic bicomponent converge to a weighted average of the synchronized trajectories of the basic bicomponents. The weights are independent of the initial conditions and are independent of the designed protocol.
comment: The article was submitted to IEEE Transactions on Automatic Control for review at March 27th, 2024
☆ Incentive-Compatible Vertiport Reservation in Advanced Air Mobility: An Auction-Based Approach
The rise of advanced air mobility (AAM) is expected to become a multibillion-dollar industry in the near future. Market-based mechanisms are touted to be an integral part of AAM operations, which comprise heterogeneous operators with private valuations. In this work, we study the problem of designing a mechanism to coordinate the movement of electric vertical take-off and landing (eVTOL) aircraft, operated by multiple operators each having heterogeneous valuations associated with their fleet, between vertiports, while enforcing the arrival, departure, and parking constraints at vertiports. Particularly, we propose an incentive-compatible and individually rational vertiport reservation mechanism that maximizes a social welfare metric, which encapsulates the objective of maximizing the overall valuations of all operators while minimizing the congestion at vertiports. Additionally, we improve the computational tractability of designing the reservation mechanism by proposing a mixed binary linear programming approach that is based on constructing network flow graph corresponding to the underlying problem.
comment: 26 pages, 2 figures, 1 table
☆ Incentive Designs for Learning Agents to Stabilize Coupled Exogenous Systems
We consider a large population of learning agents noncooperatively selecting strategies from a common set, influencing the dynamics of an exogenous system (ES) we seek to stabilize at a desired equilibrium. Our approach is to design a dynamic payoff mechanism capable of shaping the population's strategy profile, thus affecting the ES's state, by offering incentives for specific strategies within budget limits. Employing system-theoretic passivity concepts, we establish conditions under which a payoff mechanism can be systematically constructed to ensure the global asymptotic stabilization of the ES's equilibrium. In comparison to previous approaches originally studied in the context of the so-called epidemic population games, the method proposed here allows for more realistic epidemic models and other types of ES, such as predator-prey dynamics. Stabilization is established with the support of a Lyapunov function, which provides useful bounds on the transients.
comment: 8 pages, 3 figures
☆ A Study of Three Influencer Archetypes for the Control of Opinion Spread in Time-Varying Social Networks
In this work we consider the impact of information spread in time-varying social networks, where agents request to follow other agents with aligned opinions while dropping ties to neighbors whose posts are too dissimilar to their own views. Opinion control and rhetorical influence has a very long history, employing various methods including education, persuasion, propaganda, marketing, and manipulation through mis-, dis-, and mal-information. The automation of opinion controllers, however, has only recently become easily deployable at a wide scale, with the advent of large language models (LLMs) and generative AI that can translate the quantified commands from opinion controllers into actual content with the appropriate nuance. Automated agents in social networks can be deployed for various purposes, such as breaking up echo chambers, bridging valuable new connections between agents, or shaping the opinions of a target population -- and all of these raise important ethical concerns that deserve serious attention and thoughtful discussion and debate. This paper attempts to contribute to this discussion by considering three archetypal influencing styles observed by human drivers in these settings, comparing and contrasting the impact of these different control methods on the opinions of agents in the network. We will demonstrate the efficacy of current generative AI for generating nuanced content consistent with the command signal from automatic opinion controllers like these, and we will report on frameworks for approaching the relevant ethical considerations.
comment: Submission to IEEE 2024 Conference on Decision and Control. 8 pages, 7 figures, 1 table
♻ ☆ CrystalBox: Future-Based Explanations for Input-Driven Deep RL Systems
We present CrystalBox, a novel, model-agnostic, posthoc explainability framework for Deep Reinforcement Learning (DRL) controllers in the large family of input-driven environments which includes computer systems. We combine the natural decomposability of reward functions in input-driven environments with the explanatory power of decomposed returns. We propose an efficient algorithm to generate future-based explanations across both discrete and continuous control environments. Using applications such as adaptive bitrate streaming and congestion control, we demonstrate CrystalBox's capability to generate high-fidelity explanations. We further illustrate its higher utility across three practical use cases: contrastive explanations, network observability, and guided reward design, as opposed to prior explainability techniques that identify salient features.
♻ ☆ Unconstrained learning of networked nonlinear systems via free parametrization of stable interconnected operators
This paper characterizes a new parametrization of nonlinear networked incrementally $L_2$-bounded operators in discrete time. The distinctive novelty is that our parametrization is \emph{free} -- that is, a sparse large-scale operator with bounded incremental $L_2$ gain is obtained for any choice of the real values of our parameters. This property allows one to freely search over optimal parameters via unconstrained gradient descent, enabling direct applications in large-scale optimal control and system identification. Further, we can embed prior knowledge about the interconnection topology and stability properties of the system directly into the large-scale distributed operator we design. Our approach is extremely general in that it can seamlessly encapsulate and interconnect state-of-the-art Neural Network (NN) parametrizations of stable dynamical systems. To demonstrate the effectiveness of this approach, we provide a simulation example showcasing the identification of a networked nonlinear system. The results underscore the superiority of our free parametrizations over standard NN-based identification methods where a prior over the system topology and local stability properties are not enforced.
comment: Full version of the paper to appear at ECC 2024
♻ ☆ Nonlinear Control Allocation: A Learning Based Approach IEEE Conference on Decision and Control
Modern aircraft are designed with redundant control effectors to cater for fault tolerance and maneuverability requirements. This leads to aircraft being over-actuated and requires control allocation schemes to distribute the control commands among control effectors. Traditionally, optimization-based control allocation schemes are used; however, for nonlinear allocation problems, these methods require large computational resources. In this work, an artificial neural network (ANN) based nonlinear control allocation scheme is proposed. The proposed scheme is composed of learning the inverse of the control effectiveness map through ANN, and then implementing it as an allocator instead of solving an online optimization problem. Stability conditions are presented for closed-loop systems incorporating the allocator, and computational challenges are explored with piece-wise linear effectiveness functions and ANN-based allocators. To demonstrate the efficacy of the proposed scheme, it is compared with a standard quadratic programming-based method for control allocation.
comment: submitted to IEEE Conference on Decision and Control (CDC), 2024
♻ ☆ Input Convex Lipschitz RNN: A Fast and Robust Approach for Engineering Tasks
Computational efficiency and non-adversarial robustness are critical factors in real-world engineering applications. Yet, conventional neural networks often fall short in addressing both simultaneously, or even separately. Drawing insights from natural physical systems and existing literature, it is known that an input convex architecture enhances computational efficiency, while a Lipschitz-constrained architecture bolsters non-adversarial robustness. By leveraging the strengths of convexity and Lipschitz continuity, we develop a novel network architecture, termed Input Convex Lipschitz Recurrent Neural Networks. This model is explicitly designed for fast and robust optimization-based tasks and outperforms existing recurrent units across a spectrum of engineering tasks in terms of computational efficiency and non-adversarial robustness, including real-world solar irradiance prediction for Solar PV system planning at LHT Holdings in Singapore and real-time Model Predictive Control optimization for a nonlinear chemical reactor.
♻ ☆ Stochastic Approximation with Delayed Updates: Finite-Time Rates under Markovian Sampling AISTATS
Motivated by applications in large-scale and multi-agent reinforcement learning, we study the non-asymptotic performance of stochastic approximation (SA) schemes with delayed updates under Markovian sampling. While the effect of delays has been extensively studied for optimization, the manner in which they interact with the underlying Markov process to shape the finite-time performance of SA remains poorly understood. In this context, our first main contribution is to show that under time-varying bounded delays, the delayed SA update rule guarantees exponentially fast convergence of the \emph{last iterate} to a ball around the SA operator's fixed point. Notably, our bound is \emph{tight} in its dependence on both the maximum delay $\tau_{max}$, and the mixing time $\tau_{mix}$. To achieve this tight bound, we develop a novel inductive proof technique that, unlike various existing delayed-optimization analyses, relies on establishing uniform boundedness of the iterates. As such, our proof may be of independent interest. Next, to mitigate the impact of the maximum delay on the convergence rate, we provide the first finite-time analysis of a delay-adaptive SA scheme under Markovian sampling. In particular, we show that the exponent of convergence of this scheme gets scaled down by $\tau_{avg}$, as opposed to $\tau_{max}$ for the vanilla delayed SA rule; here, $\tau_{avg}$ denotes the average delay across all iterations. Moreover, the adaptive scheme requires no prior knowledge of the delay sequence for step-size tuning. Our theoretical findings shed light on the finite-time effects of delays for a broad class of algorithms, including TD learning, Q-learning, and stochastic gradient descent under Markovian sampling.
comment: Accepted to the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024!
♻ ☆ Redesigning Large-Scale Multimodal Transit Networks with Shared Autonomous Mobility Services
This study addresses a large-scale multimodal transit network design problem, with Shared Autonomous Mobility Services (SAMS) as both transit feeders and an origin-to-destination mode. The framework captures spatial demand and modal characteristics, considers intermodal transfers and express services, determines transit infrastructure investment and path flows, and generates transit routes. A system-optimal multimodal transit network is designed with minimum total door-to-door generalized costs of users and operators, satisfying transit origin-destination demand within a pre-set infrastructure budget. Firstly, the geography, demand, and modes in each zone are characterized with continuous approximation. The decisions of network link investment and multimodal path flows in zonal connection optimization are formulated as a minimum-cost multi-commodity network flow (MCNF) problem and solved efficiently with a mixed-integer linear programming (MILP) solver. Subsequently, the route generation problem is solved by expanding the MCNF formulation to minimize intramodal transfers. The model is illustrated through a set of experiments with the Chicago network comprised of 50 zones and seven modes, under three scenarios. The computational results present savings in traveler journey time and operator cost demonstrating the potential benefits of collaboration between multimodal transit systems and SAMS.
comment: 48 pages, 18 figures, accepted for publication in Transportation Research Part C: Emerging Technologies, and presentation in the 25th International Symposium on Transportation and Traffic Theory (ISTTT25)
♻ ☆ Aggregate Model of District Heating Network for Integrated Energy Dispatch: A Physically Informed Data-Driven Approach
The district heating network (DHN) is essential in enhancing the operational flexibility of integrated energy systems (IES). Yet, it is hard to obtain an accurate and concise DHN model for the operation owing to complicated network features and imperfect measurements. Considering this, this paper proposes a physical-ly informed data-driven aggregate model (AGM) for the DHN, providing a concise description of the source-load relationship of DHN without exposing network details. First, we derive the analytical relationship between the state variables of the source and load nodes of the DHN, offering a physical fundament for the AGM. Second, we propose a physics-informed estimator for the AGM that is robust to low-quality measurements, in which the physical constraints associated with the parameter normalization and sparsity are embedded to improve the accuracy and robustness. Finally, we propose a physics-enhanced algorithm to solve the nonlinear estimator with non-closed constraints efficiently. Simulation results verify the effectiveness of the proposed method.
♻ ☆ Natural-artificial hybrid swarm: Cyborg-insect group navigation in unknown obstructed soft terrain
Navigating multi-robot systems in complex terrains has always been a challenging task. This is due to the inherent limitations of traditional robots in collision avoidance, adaptation to unknown environments, and sustained energy efficiency. In order to overcome these limitations, this research proposes a solution by integrating living insects with miniature electronic controllers to enable robotic-like programmable control, and proposing a novel control algorithm for swarming. Although these creatures, called cyborg insects, have the ability to instinctively avoid collisions with neighbors and obstacles while adapting to complex terrains, there is a lack of literature on the control of multi-cyborg systems. This research gap is due to the difficulty in coordinating the movements of a cyborg system under the presence of insects' inherent individual variability in their reactions to control input. In response to this issue, we propose a novel swarm navigation algorithm addressing these challenges. The effectiveness of the algorithm is demonstrated through an experimental validation in which a cyborg swarm was successfully navigated through an unknown sandy field with obstacles and hills. This research contributes to the domain of swarm robotics and showcases the potential of integrating biological organisms with robotics and control theory to create more intelligent autonomous systems with real-world applications.
♻ ☆ Polygonal Cone Control Barrier Functions (PolyC2BF) for safe navigation in cluttered environments
In fields such as mining, search and rescue, and archaeological exploration, ensuring real-time, collision-free navigation of robots in confined, cluttered environments is imperative. Despite the value of established path planning algorithms, they often face challenges in convergence rates and handling dynamic infeasibilities. Alternative techniques like collision cones struggle to accurately represent complex obstacle geometries. This paper introduces a novel category of control barrier functions, known as Polygonal Cone Control Barrier Function (PolyC2BF), which addresses overestimation and computational complexity issues. The proposed PolyC2BF, formulated as a Quadratic Programming (QP) problem, proves effective in facilitating collision-free movement of multiple robots in complex environments. The efficacy of this approach is further demonstrated through PyBullet simulations on quadruped (unicycle model), and crazyflie 2.1 (quadrotor model) in cluttered environments.
comment: 6 Pages, 6 Figures. Accepted at European Control Conference (ECC) 2024. arXiv admin note: text overlap with arXiv:2303.15871
♻ ☆ Robust Commutation Design: Applied to Switched Reluctance Motors
Switched Reluctance Motors (SRMs) are cost-effective electric actuators that utilize magnetic reluctance to generate torque, with torque ripple arising from unaccounted manufacturing defects in the rotor tooth geometry. This paper aims to design a versatile, resource-efficient commutation function for accurate control of a range of SRMs, mitigating torque ripple despite manufacturing variations across SRMs and individual rotor teeth. The developed commutation function optimally distributes current between coils by leveraging the variance in the torque-current-angle model and is designed with few parameters for easy integration on affordable hardware. Monte Carlo simulations and experimental results show a tracking error reduction of up to 31% and 11%, respectively. The developed approach is beneficial for applications using a single driver for multiple systems and those constrained by memory or modeling effort, providing an economical solution for improved tracking performance and reduced acoustic noise.
comment: 6 pages, 7 figures. Final version
♻ ☆ Enhancing Performance, Calibration Time and Efficiency in Brain-Machine Interfaces through Transfer Learning and Wearable EEG Technology
Brain-machine interfaces (BMIs) have emerged as a transformative force in assistive technologies, empowering individuals with motor impairments by enabling device control and facilitating functional recovery. However, the persistent challenge of inter-session variability poses a significant hurdle, requiring time-consuming calibration at every new use. Compounding this issue, the low comfort level of current devices further restricts their usage. To address these challenges, we propose a comprehensive solution that combines a tiny CNN-based Transfer Learning (TL) approach with a comfortable, wearable EEG headband. The novel wearable EEG device features soft dry electrodes placed on the headband and is capable of on-board processing. We acquire multiple sessions of motor-movement EEG data and achieve up to 96% inter-session accuracy using TL, greatly reducing the calibration time and improving usability. By executing the inference on the edge every 100ms, the system is estimated to achieve 30h of battery life. The comfortable BMI setup with tiny CNN and TL paves the way to future on-device continual learning, essential for tackling inter-session variability and improving usability.
♻ ☆ Nigel -- Mechatronic Design and Robust Sim2Real Control of an Over-Actuated Autonomous Vehicle
Simulation to reality (sim2real) transfer from a dynamics and controls perspective usually involves re-tuning or adapting the designed algorithms to suit real-world operating conditions, which often violates the performance guarantees established originally. This work presents a generalizable framework for achieving reliable sim2real transfer of autonomy-oriented control systems using multi-model multi-objective robust optimal control synthesis, which lends well to uncertainty handling and disturbance rejection with theoretical guarantees. Particularly, this work is centered around a novel actuation-redundant scaled autonomous vehicle called Nigel, with independent all-wheel drive and independent all-wheel steering architecture, whose enhanced configuration space bodes well for robust control applications. To this end, we present the mechatronic design, dynamics modeling, parameter identification, and robust stabilizing as well as tracking control of Nigel using the proposed framework, with exhaustive experimentation and benchmarking in simulation as well as real-world settings.
♻ ☆ Ensuring Disturbance Rejection Performance by Synthesizing Grid-Following and Grid-Forming Inverters in Power Systems
To satisfy dynamic requirements of power systems, it is imperative for grid-tied inverters to ensure good disturbance rejection performance (DRP) under variable grid conditions. This letter discovers and theoretically proves that for general networks, synthesizing grid-following (GFL) inverters and grid-forming (GFM) inverters can always more effectively ensure the DRP of multiple inverters, as compared to homogeneous inverter-based systems that solely utilize either GFL or GFM inverters. The synthesis of GFL inverters and GFM inverters can concurrently increase the smallest eigenvalue and decrease the largest eigenvalue of the network grounded Laplacian matrix. This can be equivalent to rematching the proper short-circuit ratio (SCR) for GFL and GFM inverters, thereby ensuring the DRP of inverters both in weak and strong grids. The results reveal the necessity of synthesizing diverse inverter control schemes from the network-based perspective. Sensitivity function-based tests and real-time simulations validate our results.
comment: 6 pages
Robotics 59
☆ SLEDGE: Synthesizing Simulation Environments for Driving Agents with Generative Models
SLEDGE is the first generative simulator for vehicle motion planning trained on real-world driving logs. Its core component is a learned model that is able to generate agent bounding boxes and lane graphs. The model's outputs serve as an initial state for traffic simulation. The unique properties of the entities to be generated for SLEDGE, such as their connectivity and variable count per scene, render the naive application of most modern generative models to this task non-trivial. Therefore, together with a systematic study of existing lane graph representations, we introduce a novel raster-to-vector autoencoder (RVAE). It encodes agents and the lane graph into distinct channels in a rasterized latent map. This facilitates both lane-conditioned agent generation and combined generation of lanes and agents with a Diffusion Transformer. Using generated entities in SLEDGE enables greater control over the simulation, e.g. upsampling turns or increasing traffic density. Further, SLEDGE can support 500m long routes, a capability not found in existing data-driven simulators like nuPlan. It presents new challenges for planning algorithms, evidenced by failure rates of over 40% for PDM, the winner of the 2023 nuPlan challenge, when tested on hard routes and dense traffic generated by our model. Compared to nuPlan, SLEDGE requires 500$\times$ less storage to set up (<4GB), making it a more accessible option and helping with democratizing future research in this field.
☆ Multi-Agent Clarity-Aware Dynamic Coverage with Gaussian Processes
This paper presents two algorithms for multi-agent dynamic coverage in spatiotemporal environments, where the coverage algorithms are informed by the method of data assimilation. In particular, we show that by considering the information assimilation algorithm, here a Numerical Gaussian Process Kalman Filter, the influence of measurements taken at one position on the uncertainty of the estimate at another location can be computed. We use this relationship to propose new coverage algorithms. Furthermore, we show that the controllers naturally extend to the multi-agent context, allowing for a distributed-control central-information paradigm for multi-agent coverage. Finally, we demonstrate the algorithms through a realistic simulation of a team of UAVs collecting wind data over a region in Austria.
comment: 8 pages, 2 figures, submitted to CDC 2024
☆ CMP: Cooperative Motion Prediction with Multi-Agent Communication
The confluence of the advancement of Autonomous Vehicles (AVs) and the maturity of Vehicle-to-Everything (V2X) communication has enabled the capability of cooperative connected and automated vehicles (CAVs). Building on top of cooperative perception, this paper explores the feasibility and effectiveness of cooperative motion prediction. Our method, CMP, takes LiDAR signals as input to enhance tracking and prediction capabilities. Unlike previous work that focuses separately on either cooperative perception or motion prediction, our framework, to the best of our knowledge, is the first to address the unified problem where CAVs share information in both perception and prediction modules. Incorporated into our design is the unique capability to tolerate realistic V2X bandwidth limitations and transmission delays, while dealing with bulky perception representations. We also propose a prediction aggregation module, which unifies the predictions obtained by different CAVs and generates the final prediction. Through extensive experiments and ablation studies, we demonstrate the effectiveness of our method in cooperative perception, tracking, and motion prediction tasks. In particular, CMP reduces the average prediction error by 17.2\% with fewer missing detections compared with the no cooperation setting. Our work marks a significant step forward in the cooperative capabilities of CAVs, showcasing enhanced performance in complex scenarios.
☆ Multi Agent Pathfinding for Noise Restricted Hybrid Fuel Unmanned Aerial Vehicles
Multi Agent Path Finding (MAPF) seeks the optimal set of paths for multiple agents from respective start to goal locations such that no paths conflict. We address the MAPF problem for a fleet of hybrid-fuel unmanned aerial vehicles which are subject to location-dependent noise restrictions. We solve this problem by searching a constraint tree for which the subproblem at each node is a set of shortest path problems subject to the noise and fuel constraints and conflict zone avoidance. A labeling algorithm is presented to solve this subproblem, including the conflict zones which are treated as dynamic obstacles. We present the experimental results of the algorithms for various graph sizes and number of agents.
comment: 6 pages, 7 figures
☆ Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation
Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.
comment: Code and video are available at http://hovsg.github.io/
☆ Scenario-Based Curriculum Generation for Multi-Agent Autonomous Driving
The automated generation of diverse and complex training scenarios has been an important ingredient in many complex learning tasks. Especially in real-world application domains, such as autonomous driving, auto-curriculum generation is considered vital for obtaining robust and general policies. However, crafting traffic scenarios with multiple, heterogeneous agents is typically considered as a tedious and time-consuming task, especially in more complex simulation environments. In our work, we introduce MATS-Gym, a Multi-Agent Traffic Scenario framework to train agents in CARLA, a high-fidelity driving simulator. MATS-Gym is a multi-agent training framework for autonomous driving that uses partial scenario specifications to generate traffic scenarios with variable numbers of agents. This paper unifies various existing approaches to traffic scenario description into a single training framework and demonstrates how it can be integrated with techniques from unsupervised environment design to automate the generation of adaptive auto-curricula. The code is available at https://github.com/AutonomousDrivingExaminer/mats-gym.
comment: 7 Pages, Under Review
☆ System Calibration of a Field Phenotyping Robot with Multiple High-Precision Profile Laser Scanners
The creation of precise and high-resolution crop point clouds in agricultural fields has become a key challenge for high-throughput phenotyping applications. This work implements a novel calibration method to calibrate the laser scanning system of an agricultural field robot consisting of two industrial-grade laser scanners used for high-precise 3D crop point cloud creation. The calibration method optimizes the transformation between the scanner origins and the robot pose by minimizing 3D point omnivariances within the point cloud. Moreover, we present a novel factor graph-based pose estimation method that fuses total station prism measurements with IMU and GNSS heading information for high-precise pose determination during calibration. The root-mean-square error of the distances to a georeferenced ground truth point cloud results in 0.8 cm after parameter optimization. Furthermore, our results show the importance of a reference point cloud in the calibration method needed to estimate the vertical translation of the calibration. Challenges arise due to non-static parameters while the robot moves, indicated by systematic deviations to a ground truth terrestrial laser scan.
☆ Optical Flow Based Detection and Tracking of Moving Objects for Autonomous Vehicles
Accurate velocity estimation of surrounding moving objects and their trajectories are critical elements of perception systems in Automated/Autonomous Vehicles (AVs) with a direct impact on their safety. These are non-trivial problems due to the diverse types and sizes of such objects and their dynamic and random behaviour. Recent point cloud based solutions often use Iterative Closest Point (ICP) techniques, which are known to have certain limitations. For example, their computational costs are high due to their iterative nature, and their estimation error often deteriorates as the relative velocities of the target objects increase (>2 m/sec). Motivated by such shortcomings, this paper first proposes a novel Detection and Tracking of Moving Objects (DATMO) for AVs based on an optical flow technique, which is proven to be computationally efficient and highly accurate for such problems. \textcolor{black}{This is achieved by representing the driving scenario as a vector field and applying vector calculus theories to ensure spatiotemporal continuity.} We also report the results of a comprehensive performance evaluation of the proposed DATMO technique, carried out in this study using synthetic and real-world data. The results of this study demonstrate the superiority of the proposed technique, compared to the DATMO techniques in the literature, in terms of estimation accuracy and processing time in a wide range of relative velocities of moving objects. Finally, we evaluate and discuss the sensitivity of the estimation error of the proposed DATMO technique to various system and environmental parameters, as well as the relative velocities of the moving objects.
comment: This manuscript has been accepted as a regular paper in Transactions on Intelligent Transportation Systems (DOI: 10.1109/TITS.2024.3382495)
☆ LiDAR-Based Crop Row Detection Algorithm for Over-Canopy Autonomous Navigation in Agriculture Fields IROS 2024
Autonomous navigation is crucial for various robotics applications in agriculture. However, many existing methods depend on RTK-GPS systems, which are expensive and susceptible to poor signal coverage. This paper introduces a state-of-the-art LiDAR-based navigation system that can achieve over-canopy autonomous navigation in row-crop fields, even when the canopy fully blocks the interrow spacing. Our crop row detection algorithm can detect crop rows across diverse scenarios, encompassing various crop types, growth stages, weed presence, and discontinuities within the crop rows. Without utilizing the global localization of the robot, our navigation system can perform autonomous navigation in these challenging scenarios, detect the end of the crop rows, and navigate to the next crop row autonomously, providing a crop-agnostic approach to navigate the whole row-crop field. This navigation system has undergone tests in various simulated agricultural fields, achieving an average of $2.98cm$ autonomous driving accuracy without human intervention on the custom Amiga robot. In addition, the qualitative results of our crop row detection algorithm from the actual soybean fields validate our LiDAR-based crop row detection algorithm's potential for practical agricultural applications.
comment: 7 pages, 9 figures, submitted to IROS 2024
☆ Learning Goal-Directed Object Pushing in Cluttered Scenes with Location-Based Attention IROS
Non-prehensile planar pushing is a challenging task due to its underactuated nature with hybrid-dynamics, where a robot needs to reason about an object's long-term behaviour and contact-switching, while being robust to contact uncertainty. The presence of clutter in the environment further complicates this task, introducing the need to include more sophisticated spatial analysis to avoid collisions. Building upon prior work on reinforcement learning (RL) with multimodal categorical exploration for planar pushing, in this paper we incorporate location-based attention to enable robust navigation through clutter. Unlike previous RL literature addressing this obstacle avoidance pushing task, our framework requires no predefined global paths and considers the target orientation of the manipulated object. Our results demonstrate that the learned policies successfully navigate through a wide range of complex obstacle configurations, including dynamic obstacles, with smooth motions, achieving the desired target object pose. We also validate the transferability of the learned policies to robotic hardware using the KUKA iiwa robot arm.
comment: Submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
☆ UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps
In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains. Our code is open-source and will be available soon.
☆ Online Tree Reconstruction and Forest Inventory on a Mobile Robotic System
Terrestrial laser scanning (TLS) is the standard technique used to create accurate point clouds for digital forest inventories. However, the measurement process is demanding, requiring up to two days per hectare for data collection, significant data storage, as well as resource-heavy post-processing of 3D data. In this work, we present a real-time mapping and analysis system that enables online generation of forest inventories using mobile laser scanners that can be mounted e.g. on mobile robots. Given incrementally created and locally accurate submaps-data payloads-our approach extracts tree candidates using a custom, Voronoi-inspired clustering algorithm. Tree candidates are reconstructed using an adapted Hough algorithm, which enables robust modeling of the tree stem. Further, we explicitly incorporate the incremental nature of the data collection by consistently updating the database using a pose graph LiDAR SLAM system. This enables us to refine our estimates of the tree traits if an area is revisited later during a mission. We demonstrate competitive accuracy to TLS or manual measurements using laser scanners that we mounted on backpacks or mobile robots operating in conifer, broad-leaf and mixed forests. Our results achieve RMSE of 1.93 cm, a bias of 0.65 cm and a standard deviation of 1.81 cm (averaged across these sequences)-with no post-processing required after the mission is complete.
☆ Interactive Identification of Granular Materials using Force Measurements IROS 2024
The ability to identify granular materials facilitates the emergence of various new applications in robotics, ranging from cooking at home to truck loading at mining sites. However, granular material identification remains a challenging and underexplored area. In this work, we present a novel interactive material identification framework that enables robots to identify a wide range of granular materials using only a force-torque sensor for perception. Our framework, comprising interactive exploration, feature extraction, and classification stages, prioritizes simplicity and transparency for seamless integration into various manipulation pipelines. We evaluate the proposed approach through extensive experiments with a real-world dataset comprising 11 granular materials, which we also make publicly available. Additionally, we conducted a comprehensive qualitative analysis of the dataset to offer deeper insights into its nature, aiding future development. Our results show that the proposed method is capable of accurately identifying a wide range of granular materials solely relying on force measurements obtained from direct interaction with the materials. Code and dataset are available at: https://irobotics.aalto.fi/indentify_granular/.
comment: Submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)
☆ Aerial Robots Carrying Flexible Cables: Dynamic Shape Optimal Control via Spectral Method Model
In this work, we present a model-based optimal boundary control design for an aerial robotic system composed of a quadrotor carrying a flexible cable. The whole system is modeled by partial differential equations (PDEs) combined with boundary conditions described by ordinary differential equations (ODEs). The proper orthogonal decomposition (POD) method is adopted to project the original infinite-dimensional system on a subspace spanned by orthogonal basis functions. Based on the reduced order model, nonlinear model predictive control (NMPC) is implemented online to realize shape trajectory tracking of the flexible cable in an optimal predictive fashion. The proposed reduced modeling and optimal control paradigms are numerically verified against an accurate high-dimensional FDM-based model in different scenarios and the controller's superior performance is shown compared to an optimally tuned PID controller.
☆ Time-Optimal Flight with Safety Constraints and Data-driven Dynamics
Time-optimal quadrotor flight is an extremely challenging problem due to the limited control authority encountered at the limit of handling. Model Predictive Contouring Control (MPCC) has emerged as a leading model-based approach for time optimization problems such as drone racing. However, the standard MPCC formulation used in quadrotor racing introduces the notion of the gates directly in the cost function, creating a multi-objective optimization that continuously trades off between maximizing progress and tracking the path accurately. This paper introduces three key components that enhance the MPCC approach for drone racing. First and foremost, we provide safety guarantees in the form of a constraint and terminal set. The safety set is designed as a spatial constraint which prevents gate collisions while allowing for time-optimization only in the cost function. Second, we augment the existing first principles dynamics with a residual term that captures complex aerodynamic effects and thrust forces learned directly from real world data. Third, we use Trust Region Bayesian Optimization (TuRBO), a state of the art global Bayesian Optimization algorithm, to tune the hyperparameters of the MPC controller given a sparse reward based on lap time minimization. The proposed approach achieves similar lap times to the best state-of-the-art RL and outperforms the best time-optimal controller while satisfying constraints. In both simulation and real-world, our approach consistently prevents gate crashes with 100\% success rate, while pushing the quadrotor to its physical limit reaching speeds of more than 80km/h.
comment: 12 pages, 7 figures
☆ DeepMIF: Deep Monotonic Implicit Fields for Large-Scale LiDAR 3D Mapping
Recently, significant progress has been achieved in sensing real large-scale outdoor 3D environments, particularly by using modern acquisition equipment such as LiDAR sensors. Unfortunately, they are fundamentally limited in their ability to produce dense, complete 3D scenes. To address this issue, recent learning-based methods integrate neural implicit representations and optimizable feature grids to approximate surfaces of 3D scenes. However, naively fitting samples along raw LiDAR rays leads to noisy 3D mapping results due to the nature of sparse, conflicting LiDAR measurements. Instead, in this work we depart from fitting LiDAR data exactly, instead letting the network optimize a non-metric monotonic implicit field defined in 3D space. To fit our field, we design a learning system integrating a monotonicity loss that enables optimizing neural monotonic fields and leverages recent progress in large-scale 3D mapping. Our algorithm achieves high-quality dense 3D mapping performance as captured by multiple quantitative and perceptual measures and visual results obtained for Mai City, Newer College, and KITTI benchmarks. The code of our approach will be made publicly available.
comment: 8 pages, 6 figures
☆ Design and Preliminary Evaluation of a Torso Stabiliser for Individuals with Spinal Cord Injury
Spinal cord injuries (SCIs) generally result in sensory and mobility impairments, with torso instability being particularly debilitating. Existing torso stabilisers are often rigid and restrictive. This paper presents an early investigation into a non-restrictive 1 degree-of-freedom (DoF) mechanical torso stabiliser inspired by devices such as centrifugal clutches and seat-belt mechanisms. Firstly, the paper presents a motion-capture (MoCap) and OpenSim-based kinematic analysis of the cable-based system to understand requisite device characteristics. The simulated evaluation resulted in the cable-based device to require 55-60cm of unrestricted travel, and to lock at a threshold cable velocity of 80-100cm/sec. Next, the developed 1-DoF device is introduced. The proposed mechanical device is transparent during activities of daily living, and transitions to compliant blocking when incipient fall is detected. Prototype behaviour was then validated using a MoCap-based kinematic analysis to verify non-restrictive movement, reliable transition to blocking, and compliance of the blocking.
comment: 4 pages, 4 figures, 10 references. Submitted to IEEE EMBC 2024 conference
☆ High-Power, Flexible, Robust Hand: Development of Musculoskeletal Hand Using Machined Springs and Realization of Self-Weight Supporting Motion with Humanoid IROS2017
Human can not only support their body during standing or walking, but also support them by hand, so that they can dangle a bar and others. But most humanoid robots support their body only in the foot and they use their hand just to manipulate objects because their hands are too weak to support their body. Strong hands are supposed to enable humanoid robots to act in much broader scene. Therefore, we developed new life-size five-fingered hand that can support the body of life-size humanoid robot. It is tendon-driven and underactuated hand and actuators in forearms produce large gripping force. This hand has flexible joints using machined springs, which can be designed integrally with the attachment. Thus, it has both structural strength and impact resistance in spite of small size. As other characteristics, this hand has force sensors to measure external force and the fingers can be flexed along objects though the number of actuators to flex fingers is less than that of fingers. We installed the developed hand on musculoskeletal humanoid "Kengoro" and achieved two self-weight supporting motions: push-up motion and dangling motion.
comment: accepted at IROS2017
☆ Five-fingered Hand with Wide Range of Thumb Using Combination of Machined Springs and Variable Stiffness Joints IROS2018
Human hands can not only grasp objects of various shape and size and manipulate them in hands but also exert such a large gripping force that they can support the body in the situations such as dangling a bar and climbing a ladder. On the other hand, it is difficult for most robot hands to manage both. Therefore in this paper we developed the hand which can grasp various objects and exert large gripping force. To develop such hand, we focused on the thumb CM joint with wide range of motion and the MP joints of four fingers with the DOF of abduction and adduction. Based on the hand with large gripping force and flexibility using machined spring, we applied above mentioned joint mechanism to the hand. The thumb CM joint has wide range of motion because of the combination of three machined springs and MP joints of four fingers have variable rigidity mechanism instead of driving each joint independently in order to move joint in limited space and by limited actuators. Using the developed hand, we achieved the grasping of various objects, supporting a large load and several motions with an arm.
comment: accepted at IROS2018
☆ Adaptive Line-Of-Sight guidance law based on vector fields path following for underactuated unmanned surface vehicle
The focus of this paper is to develop a methodology that enables an unmanned surface vehicle (USV) to efficiently track a planned path. The introduction of a vector field-based adaptive line-of-sight guidance law (VFALOS) for accurate trajectory tracking and minimizing the overshoot response time during USV tracking of curved paths improves the overall line-of-sight (LOS) guidance method. These improvements contribute to faster convergence to the desired path, reduce oscillations, and can mitigate the effects of persistent external disturbances. It is shown that the proposed guidance law exhibits k-exponential stability when converging to the desired path consisting of straight and curved lines. The results in the paper show that the proposed method effectively improves the accuracy of the USV tracking the desired path while ensuring the safety of the USV work.
☆ Adaptive LiDAR-Radar Fusion for Outdoor Odometry Across Dense Smoke Conditions
Robust odometry estimation in perceptually degraded environments represents a key challenge in the field of robotics. In this paper, we propose a LiDAR-radar fusion method for robust odometry for adverse environment with LiDAR degeneracy. By comparing the LiDAR point cloud with the radar static point cloud obtained through preprocessing module, it is possible to identify instances of LiDAR degeneracy to overcome perceptual limits. We demonstrate the effectiveness of our method in challenging conditions such as dense smoke, showcasing its ability to reliably estimate odometry and identify/remove dynamic points prone to LiDAR degeneracy.
☆ Cyclic pursuit formation control for arbitrary desired shapes
A multi-agent system comprises numerous agents that autonomously make decisions to collectively accomplish tasks, drawing significant attention for their wide-ranging applications. Within this context, formation control emerges as a prominent task, wherein agents collaboratively shape and maneuver while preserving formation integrity. Our focus centers on cyclic pursuit, a method facilitating the formation of circles, ellipses, and figure-eights under the assumption that agents can only perceive the relative positions of those preceding them. However, this method's scope has been restricted to these specific shapes, leaving the feasibility of forming other shapes uncertain. In response, our study proposes a novel method based on cyclic pursuit capable of forming a broader array of shapes, enabling agents to individually shape while pursuing preceding agents, thereby extending the repertoire of achievable formations. We present two scenarios concerning the information available to agents and devise formation control methods tailored to each scenario. Through extensive simulations, we demonstrate the efficacy of our proposed method in forming multiple shapes, including those represented as Fourier series, thereby underscoring the versatility and effectiveness of our approach.
☆ Natural-artificial hybrid swarm: Cyborg-insect group navigation in unknown obstructed soft terrain
Navigating multi-robot systems in complex terrains has always been a challenging task. This is due to the inherent limitations of traditional robots in collision avoidance, adaptation to unknown environments, and sustained energy efficiency. In order to overcome these limitations, this research proposes a solution by integrating living insects with miniature electronic controllers to enable robotic-like programmable control, and proposing a novel control algorithm for swarming. Although these creatures, called cyborg insects, have the ability to instinctively avoid collisions with neighbors and obstacles while adapting to complex terrains, there is a lack of literature on the control of multi-cyborg systems. This research gap is due to the difficulty in coordinating the movements of a cyborg system under the presence of insects' inherent individual variability in their reactions to control input. In response to this issue, we propose a novel swarm navigation algorithm addressing these challenges. The effectiveness of the algorithm is demonstrated through an experimental validation in which a cyborg swarm was successfully navigated through an unknown sandy field with obstacles and hills. This research contributes to the domain of swarm robotics and showcases the potential of integrating biological organisms with robotics and control theory to create more intelligent autonomous systems with real-world applications.
☆ RoboDuet: A Framework Affording Mobile-Manipulation and Cross-Embodiment
Combining the mobility of legged robots with the manipulation skills of arms has the potential to significantly expand the operational range and enhance the capabilities of robotic systems in performing various mobile manipulation tasks. Existing approaches are confined to imprecise six degrees of freedom (DoF) manipulation and possess a limited arm workspace. In this paper, we propose a novel framework, RoboDuet, which employs two collaborative policies to realize locomotion and manipulation simultaneously, achieving whole-body control through interactions between each other. Surprisingly, going beyond the large-range pose tracking, we find that the two-policy framework may enable cross-embodiment deployment such as using different quadrupedal robots or other arms. Our experiments demonstrate that the policies trained through RoboDuet can accomplish stable gaits, agile 6D end-effector pose tracking, and zero-shot exchange of legged robots, and can be deployed in the real world to perform various mobile manipulation tasks. Our project page with demo videos is at https://locomanip-duet.github.io .
☆ Multi-Objective Trajectory Planning with Dual-Encoder
Time-jerk optimal trajectory planning is crucial in advancing robotic arms' performance in dynamic tasks. Traditional methods rely on solving complex nonlinear programming problems, bringing significant delays in generating optimized trajectories. In this paper, we propose a two-stage approach to accelerate time-jerk optimal trajectory planning. Firstly, we introduce a dual-encoder based transformer model to establish a good preliminary trajectory. This trajectory is subsequently refined through sequential quadratic programming to improve its optimality and robustness. Our approach outperforms the state-of-the-art by up to 79.72\% in reducing trajectory planning time. Compared with existing methods, our method shrinks the optimality gap with the objective function value decreasing by up to 29.9\%.
comment: 6 pages, 7 figures, conference
☆ Unified Path and Gait Planning for Safe Bipedal Robot Navigation
Safe path and gait planning are essential for bipedal robots to navigate complex real-world environments. The prevailing approaches often plan the path and gait separately in a hierarchical fashion, potentially resulting in unsafe movements due to neglecting the physical constraints of walking robots. A safety-critical path must not only avoid obstacles but also ensure that the robot's gaits are subject to its dynamic and kinematic constraints. This work presents a novel approach that unifies path planning and gait planning via a Model Predictive Control (MPC) using the Linear Inverted Pendulum (LIP) model representing bipedal locomotion. This approach considers environmental constraints, such as obstacles, and the robot's kinematics and dynamics constraints. By using discrete-time Control Barrier Functions for obstacle avoidance, our approach generates the next foot landing position, ensuring robust walking gaits and a safe navigation path within clustered environments. We validated our proposed approach in simulation using a Digit robot in 20 randomly created environments. The results demonstrate improved performance in terms of safety and robustness when compared to hierarchical path and gait planning frameworks.
comment: 8 pages
Leveraging Symmetry in RL-based Legged Locomotion Control
Model-free reinforcement learning is a promising approach for autonomously solving challenging robotics control problems, but faces exploration difficulty without information of the robot's kinematics and dynamics morphology. The under-exploration of multiple modalities with symmetric states leads to behaviors that are often unnatural and sub-optimal. This issue becomes particularly pronounced in the context of robotic systems with morphological symmetries, such as legged robots for which the resulting asymmetric and aperiodic behaviors compromise performance, robustness, and transferability to real hardware. To mitigate this challenge, we can leverage symmetry to guide and improve the exploration in policy learning via equivariance/invariance constraints. In this paper, we investigate the efficacy of two approaches to incorporate symmetry: modifying the network architectures to be strictly equivariant/invariant, and leveraging data augmentation to approximate equivariant/invariant actor-critics. We implement the methods on challenging loco-manipulation and bipedal locomotion tasks and compare with an unconstrained baseline. We find that the strictly equivariant policy consistently outperforms other methods in sample efficiency and task performance in simulation. In addition, symmetry-incorporated approaches exhibit better gait quality, higher robustness and can be deployed zero-shot in real-world experiments.
☆ Sparse-Graph-Enabled Formation Planning for Large-Scale Aerial Swarms
The formation trajectory planning using complete graphs to model collaborative constraints becomes computationally intractable as the number of drones increases due to the curse of dimensionality. To tackle this issue, this paper presents a sparse graph construction method for formation planning to realize better efficiency-performance trade-off. Firstly, a sparsification mechanism for complete graphs is designed to ensure the global rigidity of sparsified graphs, which is a necessary condition for uniquely corresponding to a geometric shape. Secondly, a good sparse graph is constructed to preserve the main structural feature of complete graphs sufficiently. Since the graph-based formation constraint is described by Laplacian matrix, the sparse graph construction problem is equivalent to submatrix selection, which has combinatorial time complexity and needs a scoring metric. Via comparative simulations, the Max-Trace matrix-revealing metric shows the promising performance. The sparse graph is integrated into the formation planning. Simulation results with 72 drones in complex environments demonstrate that when preserving 30\% connection edges, our method has comparative formation error and recovery performance w.r.t. complete graphs. Meanwhile, the planning efficiency is improved by approximate an order of magnitude. Benchmark comparisons and ablation studies are conducted to fully validate the merits of our method.
☆ Code Generation for Conic Model-Predictive Control on Microcontrollers with TinyMPC
Conic constraints appear in many important control applications like legged locomotion, robotic manipulation, and autonomous rocket landing. However, current solvers for conic optimization problems have relatively heavy computational demands in terms of both floating-point operations and memory footprint, making them impractical for use on small embedded devices. We extend TinyMPC, an open-source, high-speed solver targeting low-power embedded control applications, to handle second-order cone constraints. We also present code-generation software to enable deployment of TinyMPC on a variety of microcontrollers. We benchmark our generated code against state-of-the-art embedded QP and SOCP solvers, demonstrating a two-order-of-magnitude speed increase over ECOS while consuming less memory. Finally, we demonstrate TinyMPC's efficacy on the Crazyflie, a lightweight, resource-constrained quadrotor with fast dynamics. TinyMPC and its code-generation tools are publicly available at https://tinympc.org.
comment: Submitted to CDC, 2024. First two authors contributed equally
☆ A Real-Time Rescheduling Algorithm for Multi-robot Plan Execution ICAPS 2024
One area of research in multi-agent path finding is to determine how replanning can be efficiently achieved in the case of agents being delayed during execution. One option is to reschedule the passing order of agents, i.e., the sequence in which agents visit the same location. In response, we propose Switchable-Edge Search (SES), an A*-style algorithm designed to find optimal passing orders. We prove the optimality of SES and evaluate its efficiency via simulations. The best variant of SES takes less than 1 second for small- and medium-sized problems and runs up to 4 times faster than baselines for large-sized problems.
comment: ICAPS 2024
☆ Efficient Multi-Band Temporal Video Filter for Reducing Human-Robot Interaction
Although mobile robots have on-board sensors to perform navigation, their efficiency in completing paths can be enhanced by planning to avoid human interaction. Infrastructure cameras can capture human activity continuously for the purpose of compiling activity analytics to choose efficient times and routes. We describe a cascade temporal filtering method to efficiently extract short- and long-term activity in two time dimensions, isochronal and chronological, for use in global path planning and local navigation respectively. The temporal filter has application either independently, or, if object recognition is also required, it can be used as a pre-filter to perform activity-gating of the more computationally expensive neural network processing. For a testbed 32-camera network, we show how this hybrid approach can achieve over 8 times improvement in frames per second throughput and 6.5 times reduction of system power use. We also show how the cost map of static objects in the ROS robot software development framework is augmented with dynamic regions determined from the temporal filter.
comment: 15 pages, 5 figures, 4 tables
☆ Path Integral Control with Rollout Clustering and Dynamic Obstacles
Model Predictive Path Integral (MPPI) control has proven to be a powerful tool for the control of uncertain systems (such as systems subject to disturbances and systems with unmodeled dynamics). One important limitation of the baseline MPPI algorithm is that it does not utilize simulated trajectories to their fullest extent. For one, it assumes that the average of all trajectories weighted by their performance index will be a safe trajectory. In this paper, multiple examples are shown where the previous assumption does not hold, and a trajectory clustering technique is presented that reduces the chances of the weighted average crossing in an unsafe region. Secondly, MPPI does not account for dynamic obstacles, so the authors put forward a novel cost function that accounts for dynamic obstacles without adding significant computation time to the overall algorithm. The novel contributions proposed in this paper were evaluated with extensive simulations to demonstrate improvements upon the state-of-the-art MPPI techniques.
comment: 8 pages, 5 figures, extended version of ACC 2024 submission
☆ ShapeGrasp: Zero-Shot Task-Oriented Grasping with Large Language Models through Geometric Decomposition
Task-oriented grasping of unfamiliar objects is a necessary skill for robots in dynamic in-home environments. Inspired by the human capability to grasp such objects through intuition about their shape and structure, we present a novel zero-shot task-oriented grasping method leveraging a geometric decomposition of the target object into simple, convex shapes that we represent in a graph structure, including geometric attributes and spatial relationships. Our approach employs minimal essential information - the object's name and the intended task - to facilitate zero-shot task-oriented grasping. We utilize the commonsense reasoning capabilities of large language models to dynamically assign semantic meaning to each decomposed part and subsequently reason over the utility of each part for the intended task. Through extensive experiments on a real-world robotics platform, we demonstrate that our grasping approach's decomposition and reasoning pipeline is capable of selecting the correct part in 92% of the cases and successfully grasping the object in 82% of the tasks we evaluate. Additional videos, experiments, code, and data are available on our project website: https://shapegrasp.github.io/.
comment: 8 pages
☆ Learning Piecewise Residuals of Control Barrier Functions for Safety of Switching Systems using Multi-Output Gaussian Processes
Control barrier functions (CBFs) have recently been introduced as a systematic tool to ensure safety by establishing set invariance. When combined with a control Lyapunov function (CLF), they form a safety-critical control mechanism. However, the effectiveness of CBFs and CLFs is closely tied to the system model. In practice, model uncertainty can jeopardize safety and stability guarantees and may lead to undesirable performance. In this paper, we develop a safe learning-based control strategy for switching systems in the face of uncertainty. We focus on the case that a nominal model is available for a true underlying switching system. This uncertainty results in piecewise residuals for each switching surface, impacting the CLF and CBF constraints. We introduce a batch multi-output Gaussian process (MOGP) framework to approximate these piecewise residuals, thereby mitigating the adverse effects of uncertainty. A particular structure of the covariance function enables us to convert the MOGP-based chance constraints CLF and CBF into second-order cone constraints, which leads to a convex optimization. We analyze the feasibility of the resulting optimization and provide the necessary and sufficient conditions for feasibility. The effectiveness of the proposed strategy is validated through a simulation of a switching adaptive cruise control system.
comment: arXiv admin note: text overlap with arXiv:2403.09573
☆ SpectralWaste Dataset: Multimodal Data for Waste Sorting Automation
The increase in non-biodegradable waste is a worldwide concern. Recycling facilities play a crucial role, but their automation is hindered by the complex characteristics of waste recycling lines like clutter or object deformation. In addition, the lack of publicly available labeled data for these environments makes developing robust perception systems challenging. Our work explores the benefits of multimodal perception for object segmentation in real waste management scenarios. First, we present SpectralWaste, the first dataset collected from an operational plastic waste sorting facility that provides synchronized hyperspectral and conventional RGB images. This dataset contains labels for several categories of objects that commonly appear in sorting plants and need to be detected and separated from the main trash flow for several reasons, such as security in the management line or reuse. Additionally, we propose a pipeline employing different object segmentation architectures and evaluate the alternatives on our dataset, conducting an extensive analysis for both multimodal and unimodal alternatives. Our evaluation pays special attention to efficiency and suitability for real-time processing and demonstrates how HSI can bring a boost to RGB-only perception in these realistic industrial settings without much computational overhead.
☆ A Study on the Use of Simulation in Synthesizing Path-Following Control Policies for Autonomous Ground Robots
We report results obtained and insights gained while answering the following question: how effective is it to use a simulator to establish path following control policies for an autonomous ground robot? While the quality of the simulator conditions the answer to this question, we found that for the simulation platform used herein, producing four control policies for path planning was straightforward once a digital twin of the controlled robot was available. The control policies established in simulation and subsequently demonstrated in the real world are PID control, MPC, and two neural network (NN) based controllers. Training the two NN controllers via imitation learning was accomplished expeditiously using seven simple maneuvers: follow three circles clockwise, follow the same circles counter-clockwise, and drive straight. A test randomization process that employs random micro-simulations is used to rank the ``goodness'' of the four control policies. The policy ranking noted in simulation correlates well with the ranking observed when the control policies were tested in the real world. The simulation platform used is publicly available and BSD3-released as open source; a public Docker image is available for reproducibility studies. It contains a dynamics engine, a sensor simulator, a ROS2 bridge, and a ROS2 autonomy stack the latter employed both in the simulator and the real world experiments.
comment: 8 pages, 7 figures
☆ A Constructive Method for Designing Safe Multirate Controllers for Differentially-Flat Systems
We present a multi-rate control architecture that leverages fundamental properties of differential flatness to synthesize controllers for safety-critical nonlinear dynamical systems. We propose a two-layer architecture, where the high-level generates reference trajectories using a linear Model Predictive Controller, and the low-level tracks this reference using a feedback controller. The novelty lies in how we couple these layers, to achieve formal guarantees on recursive feasibility of the MPC problem, and safety of the nonlinear system. Furthermore, using differential flatness, we provide a constructive means to synthesize the multi-rate controller, thereby removing the need to search for suitable Lyapunov or barrier functions, or to approximately linearize/discretize nonlinear dynamics. We show the synthesized controller is a convex optimization problem, making it amenable to real-time implementations. The method is demonstrated experimentally on a ground rover and a quadruped robotic system.
comment: 6 pages, 3 figures, accepted at IEEE Control Systems Letters 2021
♻ ☆ Safe Explicable Planning
Human expectations arise from their understanding of others and the world. In the context of human-AI interaction, this understanding may not align with reality, leading to the AI agent failing to meet expectations and compromising team performance. Explicable planning, introduced as a method to bridge this gap, aims to reconcile human expectations with the agent's optimal behavior, facilitating interpretable decision-making. However, an unresolved critical issue is ensuring safety in explicable planning, as it could result in explicable behaviors that are unsafe. To address this, we propose Safe Explicable Planning (SEP), which extends the prior work to support the specification of a safety bound. The goal of SEP is to find behaviors that align with human expectations while adhering to the specified safety criterion. Our approach generalizes the consideration of multiple objectives stemming from multiple models rather than a single model, yielding a Pareto set of safe explicable policies. We present both an exact method, guaranteeing finding the Pareto set, and a more efficient greedy method that finds one of the policies in the Pareto set. Additionally, we offer approximate solutions based on state aggregation to improve scalability. We provide formal proofs that validate the desired theoretical properties of these methods. Evaluation through simulations and physical robot experiments confirms the effectiveness of our approach for safe explicable planning.
♻ ☆ Resilient source seeking with robot swarms
We present a solution for locating the source, or maximum, of an unknown scalar field using a swarm of mobile robots. Unlike relying on the traditional gradient information, the swarm determines an ascending direction to approach the source with arbitrary precision. The ascending direction is calculated from measurements of the field strength at the robot locations and their relative positions concerning the centroid. Rather than focusing on individual robots, we focus the analysis on the density of robots per unit area to guarantee a more resilient swarm, i.e., the functionality remains even if individuals go missing or are misplaced during the mission. We reinforce the robustness of the algorithm by providing sufficient conditions for the swarm shape so that the ascending direction is almost parallel to the gradient. The swarm can respond to an unexpected environment by morphing its shape and exploiting the existence of multiple ascending directions. Finally, we validate our approach numerically with hundreds of robots. The fact that a large number of robots always calculate an ascending direction compensates for the loss of individuals and mitigates issues arising from the actuator and sensor noises.
comment: 7 pages, submitted to CDC 2024
♻ ☆ Tuning-free Quasi-stiffness Control Framework of a Powered Transfemoral Prosthesis for Task-adaptive Walking RAL
Impedance-based control represents a prevalent strategy in the development of powered transfemoral prostheses. However, creating a task-adaptive, tuning-free controller that effectively generalizes across diverse locomotion modes and terrain conditions continues to be a significant challenge. This letter proposes a tuning-free and task-adaptive quasi-stiffness control framework for powered prostheses that generalizes across various walking tasks, including the torque-angle relationship reconstruction part and the quasi-stiffness controller design part. A Gaussian Process Regression (GPR) model is introduced to predict the target features of the human joint angle and torque in a new task. Subsequently, a Kernelized Movement Primitives (KMP) is employed to reconstruct the torque-angle relationship of the new task from multiple human reference trajectories and estimated target features. Based on the torque-angle relationship of the new task, a quasi-stiffness control approach is designed for a powered prosthesis. Finally, the proposed framework is validated through practical examples, including varying speeds and inclines walking tasks. Notably, the proposed framework not only aligns with but frequently surpasses the performance of a benchmark finite state machine impedance controller (FSMIC) without necessitating manual impedance tuning and has the potential to expand to variable walking tasks in daily life for the transfemoral amputees.
comment: 8 pages, 10 figures. This work has been submitted to the IEEE-RAL for possible publication
♻ ☆ Domain Randomization via Entropy Maximization ICLR 2024
Varying dynamics parameters in simulation is a popular Domain Randomization (DR) approach for overcoming the reality gap in Reinforcement Learning (RL). Nevertheless, DR heavily hinges on the choice of the sampling distribution of the dynamics parameters, since high variability is crucial to regularize the agent's behavior but notoriously leads to overly conservative policies when randomizing excessively. In this paper, we propose a novel approach to address sim-to-real transfer, which automatically shapes dynamics distributions during training in simulation without requiring real-world data. We introduce DOmain RAndomization via Entropy MaximizatiON (DORAEMON), a constrained optimization problem that directly maximizes the entropy of the training distribution while retaining generalization capabilities. In achieving this, DORAEMON gradually increases the diversity of sampled dynamics parameters as long as the probability of success of the current policy is sufficiently high. We empirically validate the consistent benefits of DORAEMON in obtaining highly adaptive and generalizable policies, i.e. solving the task at hand across the widest range of dynamics parameters, as opposed to representative baselines from the DR literature. Notably, we also demonstrate the Sim2Real applicability of DORAEMON through its successful zero-shot transfer in a robotic manipulation setup under unknown real-world parameters.
comment: Published as a conference paper at ICLR 2024. Project website at https://gabrieletiboni.github.io/doraemon/
♻ ☆ SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM
We present SGS-SLAM, the first semantic visual SLAM system based on Gaussian Splatting. It incorporates appearance, geometry, and semantic features through multi-channel optimization, addressing the oversmoothing limitations of neural implicit SLAM systems in high-quality rendering, scene understanding, and object-level geometry. We introduce a unique semantic feature loss that effectively compensates for the shortcomings of traditional depth and color losses in object optimization. Through a semantic-guided keyframe selection strategy, we prevent erroneous reconstructions caused by cumulative errors. Extensive experiments demonstrate that SGS-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, precise semantic segmentation, and object-level geometric accuracy, while ensuring real-time rendering capabilities.
♻ ☆ Towards Source-free Domain Adaptive Semantic Segmentation via Importance-aware and Prototype-contrast Learning
Domain adaptive semantic segmentation enables robust pixel-wise understanding in real-world driving scenes. Source-free domain adaptation, as a more practical technique, addresses the concerns of data privacy and storage limitations in typical unsupervised domain adaptation methods, making it especially relevant in the context of intelligent vehicles. It utilizes a well-trained source model and unlabeled target data to achieve adaptation in the target domain. However, in the absence of source data and target labels, current solutions cannot sufficiently reduce the impact of domain shift and fully leverage the information from the target data. In this paper, we propose an end-to-end source-free domain adaptation semantic segmentation method via Importance-Aware and Prototype-Contrast (IAPC) learning. The proposed IAPC framework effectively extracts domain-invariant knowledge from the well-trained source model and learns domain-specific knowledge from the unlabeled target domain. Specifically, considering the problem of domain shift in the prediction of the target domain by the source model, we put forward an importance-aware mechanism for the biased target prediction probability distribution to extract domain-invariant knowledge from the source model. We further introduce a prototype-contrast strategy, which includes a prototype-symmetric cross-entropy loss and a prototype-enhanced cross-entropy loss, to learn target intra-domain knowledge without relying on labels. A comprehensive variety of experiments on two domain adaptive semantic segmentation benchmarks demonstrates that the proposed end-to-end IAPC solution outperforms existing state-of-the-art methods. The source code is publicly available at https://github.com/yihong-97/Source-free-IAPC.
comment: Accepted to IEEE Transactions on Intelligent Vehicles (T-IV). The source code is publicly available at https://github.com/yihong-97/Source-free-IAPC
♻ ☆ When Robotics Meets Wireless Communications: An Introductory Tutorial
The importance of ground Mobile Robots (MRs) and Unmanned Aerial Vehicles (UAVs) within the research community, industry, and society is growing fast. Many of these agents are nowadays equipped with communication systems that are, in some cases, essential to successfully achieve certain tasks. In this context, we have begun to witness the development of a new interdisciplinary research field at the intersection of robotics and communications. This research field has been boosted by the intention of integrating UAVs within the 5G and 6G communication networks. This research will undoubtedly lead to many important applications in the near future. Nevertheless, one of the main obstacles to the development of this research area is that most researchers address these problems by oversimplifying either the robotics or the communications aspect. This impedes the ability of reaching the full potential of this new interdisciplinary research area. In this tutorial, we present some of the modelling tools necessary to address problems involving both robotics and communication from an interdisciplinary perspective. As an illustrative example of such problems, we focus in this tutorial on the issue of communication-aware trajectory planning.
comment: 35 pages, 192 references
♻ ☆ Motion Generation from Fine-grained Textual Descriptions
The task of text2motion is to generate human motion sequences from given textual descriptions, where the model explores diverse mappings from natural language instructions to human body movements. While most existing works are confined to coarse-grained motion descriptions, e.g., "A man squats.", fine-grained descriptions specifying movements of relevant body parts are barely explored. Models trained with coarse-grained texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure to generate motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset specializing in fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with step-by-step instructions with pseudo-code compulsory checks. Accordingly, we design a new text2motion model, FineMotionDiffuse, making full use of fine-grained textual information. Our quantitative evaluation shows that FineMotionDiffuse trained on FineHumanML3D improves FID by a large margin of 0.38, compared with competitive baselines. According to the qualitative evaluation and case study, our model outperforms MotionDiffuse in generating spatially or chronologically composite motions, by learning the implicit mappings from fine-grained descriptions to the corresponding basic motions. We release our data at https://github.com/KunhangL/finemotiondiffuse.
♻ ☆ Attention-based Estimation and Prediction of Human Intent to augment Haptic Glove aided Control of Robotic Hand
The letter focuses on Haptic Glove (HG) based control of a Robotic Hand (RH) executing in-hand manipulation of certain objects of interest. The high dimensional motion signals in HG and RH possess intrinsic variability of kinematics resulting in difficulty to establish a direct mapping of the motion signals from HG onto the RH. An estimation mechanism is proposed to quantify the motion signal acquired from the human controller in relation to the intended goal pose of the object being held by the robotic hand. A control algorithm is presented to transform the synthesized intent at the RH and allow relocation of the object to the expected goal pose. The lag in synthesis of the intent in the presence of communication delay leads to a requirement of predicting the estimated intent. We leverage an attention-based convolutional neural network encoder to predict the trajectory of intent for a certain lookahead to compensate for the delays. The proposed methodology is evaluated across objects of different shapes, mass, and materials. We present a comparative performance of the estimation and prediction mechanisms on 5G-driven real-world robotic setup against benchmark methodologies. The test-MSE in prediction of human intent is reported to yield ~ 97.3 -98.7% improvement of accuracy in comparison to LSTM-based benchmark
♻ ☆ Guessing human intentions to avoid dangerous situations in caregiving robots IROS2024
For robots to interact socially, they must interpret human intentions and anticipate their potential outcomes accurately. This is particularly important for social robots designed for human care, which may face potentially dangerous situations for people, such as unseen obstacles in their way, that should be avoided. This paper explores the Artificial Theory of Mind (ATM) approach to inferring and interpreting human intentions. We propose an algorithm that detects risky situations for humans, selecting a robot action that removes the danger in real time. We use the simulation-based approach to ATM and adopt the 'like-me' policy to assign intentions and actions to people. Using this strategy, the robot can detect and act with a high rate of success under time-constrained situations. The algorithm has been implemented as part of an existing robotics cognitive architecture and tested in simulation scenarios. Three experiments have been conducted to test the implementation's robustness, precision and real-time response, including a simulated scenario, a human-in-the-loop hybrid configuration and a real-world scenario.
comment: 8 pages, 6 figures. Submitted to IROS2024. For associated mpeg file see https://youtu.be/87UEB8P97KY
♻ ☆ Full Attitude Intelligent Controller Design of a Heliquad under Complete Failure of an Actuator
In this paper, we design a reliable Heliquad and develop an intelligent controller to handle one actuators complete failure. Heliquad is a multi-copter similar to Quadcopter, with four actuators diagonally symmetric from the center. Each actuator has two control inputs; the first input changes the propeller blades collective pitch (also called variable pitch), and the other input changes the rotation speed. For reliable operation and high torque characteristic requirement for yaw control, a cambered airfoil is used to design propeller blades. A neural network-based control allocation is designed to provide complete control authority even under a complete loss of one actuator. Nonlinear quaternion based outer loop position control, with proportional-derivative inner loop for attitude control and neural network-based control allocation is used in controller design. The proposed controller and Heliquad designs performance is evaluated using a software-in-loop simulation to track the position reference command under failure. The results clearly indicate that the Heliquad with an intelligent controller provides necessary tracking performance even under a complete loss of one actuator.
comment: 7 pages, For video go to https://indianinstituteofscience-my.sharepoint.com/:v:/g/personal/eeshank_iisc_ac_in/EcMg2uTtE91AsHDejNkb6YMBNckaXGjeh_YMzDV6sAHZAQ?e=DrRqmN
♻ ☆ Towards Massive Interaction with Generalist Robotics: A Systematic Review of XR-enabled Remote Human-Robot Interaction Systems
The rising interest of generalist robots seek to create robots with versatility to handle multiple tasks in a variety of environments, and human will interact with such robots through immersive interfaces. In the context of human-robot interaction (HRI), this survey provides an exhaustive review of the applications of extended reality (XR) technologies in the field of remote HRI. We developed a systematic search strategy based on the PRISMA methodology. From the initial 2,561 articles selected, 100 research papers that met our inclusion criteria were included. We categorized and summarized the domain in detail, delving into XR technologies, including augmented reality (AR), virtual reality (VR), and mixed reality (MR), and their applications in facilitating intuitive and effective remote control and interaction with robotic systems. The survey highlights existing articles on the application of XR technologies, user experience enhancement, and various interaction designs for XR in remote HRI, providing insights into current trends and future directions. We also identified potential gaps and opportunities for future research to improve remote HRI systems through XR technology to guide and inform future XR and robotics research.
♻ ☆ Autonomous Hook-Based Grasping and Transportation with Quadcopters
Payload grasping and transportation with quadcopters is an active research area that has rapidly developed over the last decade. To grasp a payload without human interaction, most state-of-the-art approaches apply robotic arms that are attached to the quadcopter body. However, due to the large weight and power consumption of these aerial manipulators, their agility and flight time are limited. This paper proposes a motion control and planning method for transportation with a lightweight, passive manipulator structure that consists of a hook attached to a quadrotor using a 1 DoF revolute joint. To perform payload grasping, transportation, and release, first, time-optimal reference trajectories are designed through specific waypoints to ensure the fast and reliable execution of the tasks. Then, a two-stage motion control approach is developed based on a robust geometric controller for precise and reliable reference tracking and a linear--quadratic payload regulator for rapid setpoint stabilization of the payload swing. Furthermore, stability of the closed-loop system is mathematically proven to give safety guarantee for its operation. The proposed control architecture and design are evaluated in a high-fidelity physical simulator, and also in real flight experiments, using a custom-made quadrotor--hook manipulator platform.
♻ ☆ Robustness Evaluation of Localization Techniques for Autonomous Racing
This work introduces SynPF, an MCL-based algorithm tailored for high-speed racing environments. Benchmarked against Cartographer, a state-of-the-art pose-graph SLAM algorithm, SynPF leverages synergies from previous particle-filtering methods and synthesizes them for the high-performance racing domain. Our extensive in-field evaluations reveal that while Cartographer excels under nominal conditions, it struggles when subjected to wheel-slip, a common phenomenon in a racing scenario due to varying grip levels and aggressive driving behaviour. Conversely, SynPF demonstrates robustness in these challenging conditions and a low-latency computation time of 1.25 ms on on-board computers without a GPU. Using the F1TENTH platform, a 1:10 scaled autonomous racing vehicle, this work not only highlights the vulnerabilities of existing algorithms in high-speed scenarios, tested up until 7.6 m/s, but also emphasizes the potential of SynPF as a viable alternative, especially in deteriorating odometry conditions.
comment: Accepted at the Design, Automation and Test in Europe Conference 2024 as an extended abstract
♻ ☆ OCC-VO: Dense Mapping via 3D Occupancy-Based Visual Odometry for Autonomous Driving
Visual Odometry (VO) plays a pivotal role in autonomous systems, with a principal challenge being the lack of depth information in camera images. This paper introduces OCC-VO, a novel framework that capitalizes on recent advances in deep learning to transform 2D camera images into 3D semantic occupancy, thereby circumventing the traditional need for concurrent estimation of ego poses and landmark locations. Within this framework, we utilize the TPV-Former to convert surround view cameras' images into 3D semantic occupancy. Addressing the challenges presented by this transformation, we have specifically tailored a pose estimation and mapping algorithm that incorporates Semantic Label Filter, Dynamic Object Filter, and finally, utilizes Voxel PFilter for maintaining a consistent global semantic map. Evaluations on the Occ3D-nuScenes not only showcase a 20.6% improvement in Success Ratio and a 29.6% enhancement in trajectory accuracy against ORB-SLAM3, but also emphasize our ability to construct a comprehensive map. Our implementation is open-sourced and available at: https://github.com/USTCLH/OCC-VO.
comment: 7pages, 3 figures
♻ ☆ Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models
Learning priors on trajectory distributions can help accelerate robot motion planning optimization. Given previously successful plans, learning trajectory generative models as priors for a new planning problem is highly desirable. Prior works propose several ways on utilizing this prior to bootstrapping the motion planning problem. Either sampling the prior for initializations or using the prior distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we propose learning diffusion models as priors. We then can sample directly from the posterior trajectory distribution conditioned on task goals, by leveraging the inverse denoising process of diffusion models. Furthermore, diffusion has been recently shown to effectively encode data multimodality in high-dimensional settings, which is particularly well-suited for large trajectory dataset. To demonstrate our method efficacy, we compare our proposed method - Motion Planning Diffusion - against several baselines in simulated planar robot and 7-dof robot arm manipulator environments. To assess the generalization capabilities of our method, we test it in environments with previously unseen obstacles. Our experiments show that diffusion models are strong priors to encode high-dimensional trajectory distributions of robot motions.
♻ ☆ Feeling Optimistic? Ambiguity Attitudes for Online Decision Making
Due to the complexity of many decision making problems, tree search algorithms often have inadequate information to produce accurate transition models. Robust methods, designed to make safe decisions when faced with these uncertainties, often overlook the impact expressions of uncertainty have on how the decision is made. This work introduces the Ambiguity Attitude Graph Search (AAGS), advocating for more precise representation of ambiguities (uncertainty from a set of plausible models) in decision making. Additionally, AAGS allows users to adjust their ambiguity attitude (or preference), promoting exploration and improving users' ability to control how an agent should respond when faced with a set of valid alternatives. Simulation in a dynamic sailing environment shows how highly stochastic environments can lead robust methods to fail. Results further demonstrate how adjusting ambiguity attitudes better fulfills objectives while mitigating this failure mode of robust approaches. Because this approach is a generalization of the robust framework, these results further demonstrate how algorithms focused on ambiguity have applicability beyond safety-critical systems.
comment: 6 pages, 5 figures, 2 algorithms. Submitted to the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems in Abu Dhabi, UAE (Oct 14-18, 2024)
Visual Whole-Body Control for Legged Loco-Manipulation
We study the problem of mobile manipulation using legged robots equipped with an arm, namely legged loco-manipulation. The robot legs, while usually utilized for mobility, offer an opportunity to amplify the manipulation capabilities by conducting whole-body control. That is, the robot can control the legs and the arm at the same time to extend its workspace. We propose a framework that can conduct the whole-body control autonomously with visual observations. Our approach, namely Visual Whole-Body Control(VBC), is composed of a low-level policy using all degrees of freedom to track the end-effector manipulator position and a high-level policy proposing the end-effector position based on visual inputs. We train both levels of policies in simulation and perform Sim2Real transfer for real robot deployment. We perform extensive experiments and show significant improvements over baselines in picking up diverse objects in different configurations (heights, locations, orientations) and environments. Project page: https://wholebody-b1.github.io
comment: The first two authors contribute equally. Project page: https://wholebody-b1.github.io
♻ ☆ Fast Point Cloud to Mesh Reconstruction for Deformable Object Tracking
The world around us is full of soft objects we perceive and deform with dexterous hand movements. For a robotic hand to control soft objects, it has to acquire online state feedback of the deforming object. While RGB-D cameras can collect occluded point clouds at a rate of 30Hz, this does not represent a continuously trackable object surface. Hence, in this work, we developed a method that takes as input a template mesh which is the mesh of an object in its non-deformed state and a deformed point cloud of the same object, and then shapes the template mesh such that it matches the deformed point cloud. The reconstruction of meshes from point clouds has long been studied in the field of Computer graphics under 3D reconstruction and 4D reconstruction, however, both lack the speed and generalizability needed for robotics applications. Our model is designed using a point cloud auto-encoder and a Real-NVP architecture. Our trained model can perform mesh reconstruction and tracking at a rate of 58Hz on a template mesh of 3000 vertices and a deformed point cloud of 5000 points and is generalizable to the deformations of six different object categories which are assumed to be made of soft material in our experiments (scissors, hammer, foam brick, cleanser bottle, orange, and dice). The object meshes are taken from the YCB benchmark dataset. An instance of a downstream application can be the control algorithm for a robotic hand that requires online feedback from the state of the manipulated object which would allow online grasp adaptation in a closed-loop manner. Furthermore, the tracking capacity of our method can help in the system identification of deforming objects in a marker-free approach. In future work, we will extend our trained model to generalize beyond six object categories and additionally to real-world deforming point clouds.
comment: 8 pages with appendix,16 figures
♻ ☆ Dynamic Grasping with a Learned Meta-Controller
Grasping moving objects is a challenging task that requires multiple submodules such as object pose predictor, arm motion planner, etc. Each submodule operates under its own set of meta-parameters. For example, how far the pose predictor should look into the future (i.e., look-ahead time) and the maximum amount of time the motion planner can spend planning a motion (i.e., time budget). Many previous works assign fixed values to these parameters; however, at different moments within a single episode of dynamic grasping, the optimal values should vary depending on the current scene. In this work, we propose a dynamic grasping pipeline with a meta-controller that controls the look-ahead time and time budget dynamically. We learn the meta-controller through reinforcement learning with a sparse reward. Our experiments show the meta-controller improves the grasping success rate (up to 28% in the most cluttered environment) and reduces grasping time, compared to the strongest baseline. Our meta-controller learns to reason about the reachable workspace and maintain the predicted pose within the reachable region. In addition, it assigns a small but sufficient time budget for the motion planner. Our method can handle different objects, trajectories, and obstacles. Despite being trained only with 3-6 random cuboidal obstacles, our meta-controller generalizes well to 7-9 obstacles and more realistic out-of-domain household setups with unseen obstacle shapes.
comment: 9 pages
♻ ☆ SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM CVPR 2024
Dense simultaneous localization and mapping (SLAM) is crucial for robotics and augmented reality applications. However, current methods are often hampered by the non-volumetric or implicit way they represent a scene. This work introduces SplaTAM, an approach that, for the first time, leverages explicit volumetric representations, i.e., 3D Gaussians, to enable high-fidelity reconstruction from a single unposed RGB-D camera, surpassing the capabilities of existing methods. SplaTAM employs a simple online tracking and mapping system tailored to the underlying Gaussian representation. It utilizes a silhouette mask to elegantly capture the presence of scene density. This combination enables several benefits over prior representations, including fast rendering and dense optimization, quickly determining if areas have been previously mapped, and structured map expansion by adding more Gaussians. Extensive experiments show that SplaTAM achieves up to 2x superior performance in camera pose estimation, map construction, and novel-view synthesis over existing methods, paving the way for more immersive high-fidelity SLAM applications.
comment: CVPR 2024. Website: https://spla-tam.github.io/
FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects
We present FoundationPose, a unified foundation model for 6D object pose estimation and tracking, supporting both model-based and model-free setups. Our approach can be instantly applied at test-time to a novel object without fine-tuning, as long as its CAD model is given, or a small number of reference images are captured. We bridge the gap between these two setups with a neural implicit representation that allows for effective novel view synthesis, keeping the downstream pose estimation modules invariant under the same unified framework. Strong generalizability is achieved via large-scale synthetic training, aided by a large language model (LLM), a novel transformer-based architecture, and contrastive learning formulation. Extensive evaluation on multiple public datasets involving challenging scenarios and objects indicate our unified approach outperforms existing methods specialized for each task by a large margin. In addition, it even achieves comparable results to instance-level methods despite the reduced assumptions. Project page: https://nvlabs.github.io/FoundationPose/
Artificial Intelligence 161
☆ SLEDGE: Synthesizing Simulation Environments for Driving Agents with Generative Models
SLEDGE is the first generative simulator for vehicle motion planning trained on real-world driving logs. Its core component is a learned model that is able to generate agent bounding boxes and lane graphs. The model's outputs serve as an initial state for traffic simulation. The unique properties of the entities to be generated for SLEDGE, such as their connectivity and variable count per scene, render the naive application of most modern generative models to this task non-trivial. Therefore, together with a systematic study of existing lane graph representations, we introduce a novel raster-to-vector autoencoder (RVAE). It encodes agents and the lane graph into distinct channels in a rasterized latent map. This facilitates both lane-conditioned agent generation and combined generation of lanes and agents with a Diffusion Transformer. Using generated entities in SLEDGE enables greater control over the simulation, e.g. upsampling turns or increasing traffic density. Further, SLEDGE can support 500m long routes, a capability not found in existing data-driven simulators like nuPlan. It presents new challenges for planning algorithms, evidenced by failure rates of over 40% for PDM, the winner of the 2023 nuPlan challenge, when tested on hard routes and dense traffic generated by our model. Compared to nuPlan, SLEDGE requires 500$\times$ less storage to set up (<4GB), making it a more accessible option and helping with democratizing future research in this field.
☆ MAGIS: LLM-Based Multi-Agent Framework for GitHub Issue Resolution
In software evolution, resolving the emergent issues within GitHub repositories is a complex challenge that involves not only the incorporation of new code but also the maintenance of existing functionalities. Large Language Models (LLMs) have shown promise in code generation and understanding but face difficulties in code change, particularly at the repository level. To overcome these challenges, we empirically study the reason why LLMs mostly fail to resolve GitHub issues and analyze some impact factors. Motivated by the empirical findings, we propose a novel LLM-based Multi-Agent framework for GitHub Issue reSolution, MAGIS, consisting of four kinds of agents customized for the software evolution: Manager, Repository Custodian, Developer, and Quality Assurance Engineer agents. This framework leverages the collaboration of various agents in the planning and coding process to unlock the potential of LLMs to resolve GitHub issues. In experiments, we employ the SWE-bench benchmark to compare MAGIS with popular LLMs, including GPT-3.5, GPT-4, and Claude-2. MAGIS can resolve 13.94% GitHub issues, which significantly outperforms the baselines. Specifically, MAGIS achieves an eight-fold increase in resolved ratio over the direct application of GPT-4, the based LLM of our method. We also analyze the factors for improving GitHub issue resolution rates, such as line location, task allocation, etc.
comment: work in progress
☆ AID: Attention Interpolation of Text-to-Image Diffusion
Conditional diffusion models can create unseen images in various settings, aiding image interpolation. Interpolation in latent spaces is well-studied, but interpolation with specific conditions like text or poses is less understood. Simple approaches, such as linear interpolation in the space of conditions, often result in images that lack consistency, smoothness, and fidelity. To that end, we introduce a novel training-free technique named Attention Interpolation via Diffusion (AID). Our key contributions include 1) proposing an inner/outer interpolated attention layer; 2) fusing the interpolated attention with self-attention to boost fidelity; and 3) applying beta distribution to selection to increase smoothness. We also present a variant, Prompt-guided Attention Interpolation via Diffusion (PAID), that considers interpolation as a condition-dependent generative process. This method enables the creation of new images with greater consistency, smoothness, and efficiency, and offers control over the exact path of interpolation. Our approach demonstrates effectiveness for conceptual and spatial interpolation. Code and demo are available at https://github.com/QY-H00/attention-interpolation-diffusion.
☆ LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning
The machine learning community has witnessed impressive advancements since the first appearance of large language models (LLMs), yet their huge memory consumption has become a major roadblock to large-scale training. Parameter Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA) have been proposed to alleviate this problem, but their performance still fails to match full parameter training in most large-scale fine-tuning settings. Attempting to complement this deficiency, we investigate layerwise properties of LoRA on fine-tuning tasks and observe an uncommon skewness of weight norms across different layers. Utilizing this key observation, a surprisingly simple training strategy is discovered, which outperforms both LoRA and full parameter training in a wide range of settings with memory costs as low as LoRA. We name it Layerwise Importance Sampled AdamW (LISA), a promising alternative for LoRA, which applies the idea of importance sampling to different layers in LLMs and randomly freeze most middle layers during optimization. Experimental results show that with similar or less GPU memory consumption, LISA surpasses LoRA or even full parameter tuning in downstream fine-tuning tasks, where LISA consistently outperforms LoRA by over $11\%$-$37\%$ in terms of MT-Bench scores. On large models, specifically LLaMA-2-70B, LISA achieves on-par or better performance than LoRA on MT-Bench, GSM8K, and PubMedQA, demonstrating its effectiveness across different domains.
☆ AgentStudio: A Toolkit for Building General Virtual Agents
Creating autonomous virtual agents capable of using arbitrary software on any digital device remains a major challenge for artificial intelligence. Two key obstacles hinder progress: insufficient infrastructure for building virtual agents in real-world environments, and the need for in-the-wild evaluation of fundamental agent abilities. To address this, we introduce AgentStudio, an online, realistic, and multimodal toolkit that covers the entire lifecycle of agent development. This includes environment setups, data collection, agent evaluation, and visualization. The observation and action spaces are highly generic, supporting both function calling and human-computer interfaces. This versatility is further enhanced by AgentStudio's graphical user interfaces, which allow efficient development of datasets and benchmarks in real-world settings. To illustrate, we introduce a visual grounding dataset and a real-world benchmark suite, both created with our graphical interfaces. Furthermore, we present several actionable insights derived from AgentStudio, e.g., general visual grounding, open-ended tool creation, learning from videos, etc. We have open-sourced the environments, datasets, benchmarks, and interfaces to promote research towards developing general virtual agents for the future.
☆ CMP: Cooperative Motion Prediction with Multi-Agent Communication
The confluence of the advancement of Autonomous Vehicles (AVs) and the maturity of Vehicle-to-Everything (V2X) communication has enabled the capability of cooperative connected and automated vehicles (CAVs). Building on top of cooperative perception, this paper explores the feasibility and effectiveness of cooperative motion prediction. Our method, CMP, takes LiDAR signals as input to enhance tracking and prediction capabilities. Unlike previous work that focuses separately on either cooperative perception or motion prediction, our framework, to the best of our knowledge, is the first to address the unified problem where CAVs share information in both perception and prediction modules. Incorporated into our design is the unique capability to tolerate realistic V2X bandwidth limitations and transmission delays, while dealing with bulky perception representations. We also propose a prediction aggregation module, which unifies the predictions obtained by different CAVs and generates the final prediction. Through extensive experiments and ablation studies, we demonstrate the effectiveness of our method in cooperative perception, tracking, and motion prediction tasks. In particular, CMP reduces the average prediction error by 17.2\% with fewer missing detections compared with the no cooperation setting. Our work marks a significant step forward in the cooperative capabilities of CAVs, showcasing enhanced performance in complex scenarios.
☆ Hierarchical Multi-label Classification for Fine-level Event Extraction from Aviation Accident Reports
A large volume of accident reports is recorded in the aviation domain, which greatly values improving aviation safety. To better use those reports, we need to understand the most important events or impact factors according to the accident reports. However, the increasing number of accident reports requires large efforts from domain experts to label those reports. In order to make the labeling process more efficient, many researchers have started developing algorithms to identify the underlying events from accident reports automatically. This article argues that we can identify the events more accurately by leveraging the event taxonomy. More specifically, we consider the problem a hierarchical classification task where we first identify the coarse-level information and then predict the fine-level information. We achieve this hierarchical classification process by incorporating a novel hierarchical attention module into BERT. To further utilize the information from event taxonomy, we regularize the proposed model according to the relationship and distribution among labels. The effectiveness of our framework is evaluated with the data collected by National Transportation Safety Board (NTSB). It has been shown that fine-level prediction accuracy is highly improved, and the regularization term can be beneficial to the rare event identification problem.
comment: Accepted in INFORMS Journal of Data Science
☆ Image-based Novel Fault Detection with Deep Learning Classifiers using Hierarchical Labels
One important characteristic of modern fault classification systems is the ability to flag the system when faced with previously unseen fault types. This work considers the unknown fault detection capabilities of deep neural network-based fault classifiers. Specifically, we propose a methodology on how, when available, labels regarding the fault taxonomy can be used to increase unknown fault detection performance without sacrificing model performance. To achieve this, we propose to utilize soft label techniques to improve the state-of-the-art deep novel fault detection techniques during the training process and novel hierarchically consistent detection statistics for online novel fault detection. Finally, we demonstrated increased detection performance on novel fault detection in inspection images from the hot steel rolling process, with results well replicated across multiple scenarios and baseline detection methods.
comment: Accepted in IISE Transaction
☆ Addressing Social Misattributions of Large Language Models: An HCXAI-based Approach
Human-centered explainable AI (HCXAI) advocates for the integration of social aspects into AI explanations. Central to the HCXAI discourse is the Social Transparency (ST) framework, which aims to make the socio-organizational context of AI systems accessible to their users. In this work, we suggest extending the ST framework to address the risks of social misattributions in Large Language Models (LLMs), particularly in sensitive areas like mental health. In fact LLMs, which are remarkably capable of simulating roles and personas, may lead to mismatches between designers' intentions and users' perceptions of social attributes, risking to promote emotional manipulation and dangerous behaviors, cases of epistemic injustice, and unwarranted trust. To address these issues, we propose enhancing the ST framework with a fifth 'W-question' to clarify the specific social attributions assigned to LLMs by its designers and users. This addition aims to bridge the gap between LLM capabilities and user perceptions, promoting the ethically responsible development and use of LLM-based technology.
comment: Extended version of the manuscript accepted for the ACM CHI Workshop on Human-Centered Explainable AI 2024 (HCXAI24)
☆ Climate Downscaling: A Deep-Learning Based Super-resolution Model of Precipitation Data with Attention Block and Skip Connections
Human activities accelerate consumption of fossil fuels and produce greenhouse gases, resulting in urgent issues today: global warming and the climate change. These indirectly cause severe natural disasters, plenty of lives suffering and huge losses of agricultural properties. To mitigate impacts on our lands, scientists are developing renewable, reusable, and clean energies and climatologists are trying to predict the extremes. Meanwhile, governments are publicizing resource-saving policies for a more eco-friendly society and arousing environment awareness. One of the most influencing factors is the precipitation, bringing condensed water vapor onto lands. Water resources are the most significant but basic needs in society, not only supporting our livings, but also economics. In Taiwan, although the average annual precipitation is up to 2,500 millimeter (mm), the water allocation for each person is lower than the global average due to drastically geographical elevation changes and uneven distribution through the year. Thus, it is crucial to track and predict the rainfall to make the most use of it and to prevent the floods. However, climate models have limited resolution and require intensive computational power for local-scale use. Therefore, we proposed a deep convolutional neural network with skip connections, attention blocks, and auxiliary data concatenation, in order to downscale the low-resolution precipitation data into high-resolution one. Eventually, we compare with other climate downscaling methods and show better performance in metrics of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Pearson Correlation, structural similarity index (SSIM), and forecast indicators.
☆ Hierarchical Open-Vocabulary 3D Scene Graphs for Language-Grounded Robot Navigation
Recent open-vocabulary robot mapping methods enrich dense geometric maps with pre-trained visual-language features. While these maps allow for the prediction of point-wise saliency maps when queried for a certain language concept, large-scale environments and abstract queries beyond the object level still pose a considerable hurdle, ultimately limiting language-grounded robotic navigation. In this work, we present HOV-SG, a hierarchical open-vocabulary 3D scene graph mapping approach for language-grounded robot navigation. Leveraging open-vocabulary vision foundation models, we first obtain state-of-the-art open-vocabulary segment-level maps in 3D and subsequently construct a 3D scene graph hierarchy consisting of floor, room, and object concepts, each enriched with open-vocabulary features. Our approach is able to represent multi-story buildings and allows robotic traversal of those using a cross-floor Voronoi graph. HOV-SG is evaluated on three distinct datasets and surpasses previous baselines in open-vocabulary semantic accuracy on the object, room, and floor level while producing a 75% reduction in representation size compared to dense open-vocabulary maps. In order to prove the efficacy and generalization capabilities of HOV-SG, we showcase successful long-horizon language-conditioned robot navigation within real-world multi-storage environments. We provide code and trial video data at http://hovsg.github.io/.
comment: Code and video are available at http://hovsg.github.io/
☆ ReMamber: Referring Image Segmentation with Mamba Twister
Referring Image Segmentation (RIS) leveraging transformers has achieved great success on the interpretation of complex visual-language tasks. However, the quadratic computation cost makes it resource-consuming in capturing long-range visual-language dependencies. Fortunately, Mamba addresses this with efficient linear complexity in processing. However, directly applying Mamba to multi-modal interactions presents challenges, primarily due to inadequate channel interactions for the effective fusion of multi-modal data. In this paper, we propose ReMamber, a novel RIS architecture that integrates the power of Mamba with a multi-modal Mamba Twister block. The Mamba Twister explicitly models image-text interaction, and fuses textual and visual features through its unique channel and spatial twisting mechanism. We achieve the state-of-the-art on three challenging benchmarks. Moreover, we conduct thorough analyses of ReMamber and discuss other fusion designs using Mamba. These provide valuable perspectives for future research.
☆ DiffH2O: Diffusion-Based Synthesis of Hand-Object Interactions from Textual Descriptions
Generating natural hand-object interactions in 3D is challenging as the resulting hand and object motions are expected to be physically plausible and semantically meaningful. Furthermore, generalization to unseen objects is hindered by the limited scale of available hand-object interaction datasets. We propose DiffH2O, a novel method to synthesize realistic, one or two-handed object interactions from provided text prompts and geometry of the object. The method introduces three techniques that enable effective learning from limited data. First, we decompose the task into a grasping stage and a text-based interaction stage and use separate diffusion models for each. In the grasping stage, the model only generates hand motions, whereas in the interaction phase both hand and object poses are synthesized. Second, we propose a compact representation that tightly couples hand and object poses. Third, we propose two different guidance schemes to allow more control of the generated motions: grasp guidance and detailed textual guidance. Grasp guidance takes a single target grasping pose and guides the diffusion model to reach this grasp at the end of the grasping stage, which provides control over the grasping pose. Given a grasping motion from this stage, multiple different actions can be prompted in the interaction phase. For textual guidance, we contribute comprehensive text descriptions to the GRAB dataset and show that they enable our method to have more fine-grained control over hand-object interactions. Our quantitative and qualitative evaluation demonstrates that the proposed method outperforms baseline methods and leads to natural hand-object motions. Moreover, we demonstrate the practicality of our framework by utilizing a hand pose estimate from an off-the-shelf pose estimator for guidance, and then sampling multiple different actions in the interaction stage.
comment: Project Page: https://diffh2o.github.io/
☆ On the Computational Complexity of Stackelberg Planning and Meta-Operator Verification: Technical Report ICAPS24
Stackelberg planning is a recently introduced single-turn two-player adversarial planning model, where two players are acting in a joint classical planning task, the objective of the first player being hampering the second player from achieving its goal. This places the Stackelberg planning problem somewhere between classical planning and general combinatorial two-player games. But, where exactly? All investigations of Stackelberg planning so far focused on practical aspects. We close this gap by conducting the first theoretical complexity analysis of Stackelberg planning. We show that in general Stackelberg planning is actually no harder than classical planning. Under a polynomial plan-length restriction, however, Stackelberg planning is a level higher up in the polynomial complexity hierarchy, suggesting that compilations into classical planning come with a worst-case exponential plan-length increase. In attempts to identify tractable fragments, we further study its complexity under various planning task restrictions, showing that Stackelberg planning remains intractable where classical planning is not. We finally inspect the complexity of meta-operator verification, a problem that has been recently connected to Stackelberg planning.
comment: Presented at ICAPS24
☆ Accelerating Radio Spectrum Regulation Workflows with Large Language Models (LLMs)
Wireless spectrum regulation is a complex and demanding process due to the rapid pace of technological progress, increasing demand for spectrum, and a multitude of stakeholders with potentially conflicting interests, alongside significant economic implications. To navigate this, regulators must engage effectively with all parties, keep pace with global technology trends, conduct technical evaluations, issue licenses in a timely manner, and comply with various legal and policy frameworks. In light of these challenges, this paper demonstrates example applications of Large Language Models (LLMs) to expedite spectrum regulatory processes. We explore various roles that LLMs can play in this context while identifying some of the challenges to address. The paper also offers practical case studies and insights, with appropriate experiments, highlighting the transformative potential of LLMs in spectrum management.
☆ D-PAD: Deep-Shallow Multi-Frequency Patterns Disentangling for Time Series Forecasting
In time series forecasting, effectively disentangling intricate temporal patterns is crucial. While recent works endeavor to combine decomposition techniques with deep learning, multiple frequencies may still be mixed in the decomposed components, e.g., trend and seasonal. Furthermore, frequency domain analysis methods, e.g., Fourier and wavelet transforms, have limitations in resolution in the time domain and adaptability. In this paper, we propose D-PAD, a deep-shallow multi-frequency patterns disentangling neural network for time series forecasting. Specifically, a multi-component decomposing (MCD) block is introduced to decompose the series into components with different frequency ranges, corresponding to the "shallow" aspect. A decomposition-reconstruction-decomposition (D-R-D) module is proposed to progressively extract the information of frequencies mixed in the components, corresponding to the "deep" aspect. After that, an interaction and fusion (IF) module is used to further analyze the components. Extensive experiments on seven real-world datasets demonstrate that D-PAD achieves the state-of-the-art performance, outperforming the best baseline by an average of 9.48% and 7.15% in MSE and MAE, respectively.
☆ Evaluating the Efficacy of Prompt-Engineered Large Multimodal Models Versus Fine-Tuned Vision Transformers in Image-Based Security Applications
The success of Large Language Models (LLMs) has led to a parallel rise in the development of Large Multimodal Models (LMMs), such as Gemini-pro, which have begun to transform a variety of applications. These sophisticated multimodal models are designed to interpret and analyze complex data, integrating both textual and visual information on a scale previously unattainable, opening new avenues for a range of applications. This paper investigates the applicability and effectiveness of prompt-engineered Gemini-pro LMMs versus fine-tuned Vision Transformer (ViT) models in addressing critical security challenges. We focus on two distinct tasks: a visually evident task of detecting simple triggers, such as small squares in images, indicative of potential backdoors, and a non-visually evident task of malware classification through visual representations. Our results highlight a significant divergence in performance, with Gemini-pro falling short in accuracy and reliability when compared to fine-tuned ViT models. The ViT models, on the other hand, demonstrate exceptional accuracy, achieving near-perfect performance on both tasks. This study not only showcases the strengths and limitations of prompt-engineered LMMs in cybersecurity applications but also emphasizes the unmatched efficacy of fine-tuned ViT models for precise and dependable tasks.
☆ SciCapenter: Supporting Caption Composition for Scientific Figures with Machine-Generated Captions and Ratings
Crafting effective captions for figures is important. Readers heavily depend on these captions to grasp the figure's message. However, despite a well-developed set of AI technologies for figures and captions, these have rarely been tested for usefulness in aiding caption writing. This paper introduces SciCapenter, an interactive system that puts together cutting-edge AI technologies for scientific figure captions to aid caption composition. SciCapenter generates a variety of captions for each figure in a scholarly article, providing scores and a comprehensive checklist to assess caption quality across multiple critical aspects, such as helpfulness, OCR mention, key takeaways, and visual properties reference. Users can directly edit captions in SciCapenter, resubmit for revised evaluations, and iteratively refine them. A user study with Ph.D. students indicates that SciCapenter significantly lowers the cognitive load of caption writing. Participants' feedback further offers valuable design insights for future systems aiming to enhance caption writing.
comment: CHI EA '24: Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems
☆ Towards a FAIR Documentation of Workflows and Models in Applied Mathematics
Modeling-Simulation-Optimization workflows play a fundamental role in applied mathematics. The Mathematical Research Data Initiative, MaRDI, responded to this by developing a FAIR and machine-interpretable template for a comprehensive documentation of such workflows. MaRDMO, a Plugin for the Research Data Management Organiser, enables scientists from diverse fields to document and publish their workflows on the MaRDI Portal seamlessly using the MaRDI template. Central to these workflows are mathematical models. MaRDI addresses them with the MathModDB ontology, offering a structured formal model description. Here, we showcase the interaction between MaRDMO and the MathModDB Knowledge Graph through an algebraic modeling workflow from the Digital Humanities. This demonstration underscores the versatility of both services beyond their original numerical domain.
☆ SciNews: From Scholarly Complexities to Public Narratives -- A Dataset for Scientific News Report Generation LREC
Scientific news reports serve as a bridge, adeptly translating complex research articles into reports that resonate with the broader public. The automated generation of such narratives enhances the accessibility of scholarly insights. In this paper, we present a new corpus to facilitate this paradigm development. Our corpus comprises a parallel compilation of academic publications and their corresponding scientific news reports across nine disciplines. To demonstrate the utility and reliability of our dataset, we conduct an extensive analysis, highlighting the divergences in readability and brevity between scientific news narratives and academic manuscripts. We benchmark our dataset employing state-of-the-art text generation models. The evaluation process involves both automatic and human evaluation, which lays the groundwork for future explorations into the automated generation of scientific news reports. The dataset and code related to this work are available at https://dongqi.me/projects/SciNews.
comment: LREC-COLING 2024 Main Conference Paper
☆ DataCook: Crafting Anti-Adversarial Examples for Healthcare Data Copyright Protection
In the realm of healthcare, the challenges of copyright protection and unauthorized third-party misuse are increasingly significant. Traditional methods for data copyright protection are applied prior to data distribution, implying that models trained on these data become uncontrollable. This paper introduces a novel approach, named DataCook, designed to safeguard the copyright of healthcare data during the deployment phase. DataCook operates by "cooking" the raw data before distribution, enabling the development of models that perform normally on this processed data. However, during the deployment phase, the original test data must be also "cooked" through DataCook to ensure normal model performance. This process grants copyright holders control over authorization during the deployment phase. The mechanism behind DataCook is by crafting anti-adversarial examples (AntiAdv), which are designed to enhance model confidence, as opposed to standard adversarial examples (Adv) that aim to confuse models. Similar to Adv, AntiAdv introduces imperceptible perturbations, ensuring that the data processed by DataCook remains easily understandable. We conducted extensive experiments on MedMNIST datasets, encompassing both 2D/3D data and the high-resolution variants. The outcomes indicate that DataCook effectively meets its objectives, preventing models trained on AntiAdv from analyzing unauthorized data effectively, without compromising the validity and accuracy of the data in legitimate scenarios. Code and data are available at https://github.com/MedMNIST/DataCook.
☆ Using Stratified Sampling to Improve LIME Image Explanations
We investigate the use of a stratified sampling approach for LIME Image, a popular model-agnostic explainable AI method for computer vision tasks, in order to reduce the artifacts generated by typical Monte Carlo sampling. Such artifacts are due to the undersampling of the dependent variable in the synthetic neighborhood around the image being explained, which may result in inadequate explanations due to the impossibility of fitting a linear regressor on the sampled data. We then highlight a connection with the Shapley theory, where similar arguments about undersampling and sample relevance were suggested in the past. We derive all the formulas and adjustment factors required for an unbiased stratified sampling estimator. Experiments show the efficacy of the proposed approach.
☆ All-in-One: Heterogeneous Interaction Modeling for Cold-Start Rating Prediction
Cold-start rating prediction is a fundamental problem in recommender systems that has been extensively studied. Many methods have been proposed that exploit explicit relations among existing data, such as collaborative filtering, social recommendations and heterogeneous information network, to alleviate the data insufficiency issue for cold-start users and items. However, the explicit relations constructed based on data between different roles may be unreliable and irrelevant, which limits the performance ceiling of the specific recommendation task. Motivated by this, in this paper, we propose a flexible framework dubbed heterogeneous interaction rating network (HIRE). HIRE dose not solely rely on the pre-defined interaction pattern or the manually constructed heterogeneous information network. Instead, we devise a Heterogeneous Interaction Module (HIM) to jointly model the heterogeneous interactions and directly infer the important interactions via the observed data. In the experiments, we evaluate our model under three cold-start settings on three real-world datasets. The experimental results show that HIRE outperforms other baselines by a large margin. Furthermore, we visualize the inferred interactions of HIRE to confirm the contribution of our model.
comment: 14 pages, 9 figures
☆ Out-of-distribution Rumor Detection via Test-Time Adaptation
Due to the rapid spread of rumors on social media, rumor detection has become an extremely important challenge. Existing methods for rumor detection have achieved good performance, as they have collected enough corpus from the same data distribution for model training. However, significant distribution shifts between the training data and real-world test data occur due to differences in news topics, social media platforms, languages and the variance in propagation scale caused by news popularity. This leads to a substantial decline in the performance of these existing methods in Out-Of-Distribution (OOD) situations. To address this problem, we propose a simple and efficient method named Test-time Adaptation for Rumor Detection under distribution shifts (TARD). This method models the propagation of news in the form of a propagation graph, and builds propagation graph test-time adaptation framework, enhancing the model's adaptability and robustness when facing OOD problems. Extensive experiments conducted on two group datasets collected from real-world social platforms demonstrate that our framework outperforms the state-of-the-art methods in performance.
☆ Tiny Models are the Computational Saver for Large Models
This paper introduces TinySaver, an early-exit-like dynamic model compression approach which employs tiny models to substitute large models adaptively. Distinct from traditional compression techniques, dynamic methods like TinySaver can leverage the difficulty differences to allow certain inputs to complete their inference processes early, thereby conserving computational resources. Most existing early exit designs are implemented by attaching additional network branches to the model's backbone. Our study, however, reveals that completely independent tiny models can replace a substantial portion of the larger models' job with minimal impact on performance. Employing them as the first exit can remarkably enhance computational efficiency. By searching and employing the most appropriate tiny model as the computational saver for a given large model, the proposed approaches work as a novel and generic method to model compression. This finding will help the research community in exploring new compression methods to address the escalating computational demands posed by rapidly evolving AI models. Our evaluation of this approach in ImageNet-1k classification demonstrates its potential to reduce the number of compute operations by up to 90%, with only negligible losses in performance, across various modern vision models. The code of this work will be available.
Optimization-based Prompt Injection Attack to LLM-as-a-Judge
LLM-as-a-Judge is a novel solution that can assess textual information with large language models (LLMs). Based on existing research studies, LLMs demonstrate remarkable performance in providing a compelling alternative to traditional human assessment. However, the robustness of these systems against prompt injection attacks remains an open question. In this work, we introduce JudgeDeceiver, a novel optimization-based prompt injection attack tailored to LLM-as-a-Judge. Our method formulates a precise optimization objective for attacking the decision-making process of LLM-as-a-Judge and utilizes an optimization algorithm to efficiently automate the generation of adversarial sequences, achieving targeted and effective manipulation of model evaluations. Compared to handcraft prompt injection attacks, our method demonstrates superior efficacy, posing a significant challenge to the current security paradigms of LLM-based judgment systems. Through extensive experiments, we showcase the capability of JudgeDeceiver in altering decision outcomes across various cases, highlighting the vulnerability of LLM-as-a-Judge systems to the optimization-based prompt injection attack.
☆ Enhanced Short Text Modeling: Leveraging Large Language Models for Topic Refinement
Crafting effective topic models for brief texts, like tweets and news headlines, is essential for capturing the swift shifts in social dynamics. Traditional topic models, however, often fall short in accurately representing the semantic intricacies of short texts due to their brevity and lack of contextual data. In our study, we harness the advanced capabilities of Large Language Models (LLMs) to introduce a novel approach termed "Topic Refinement". This approach does not directly involve itself in the initial modeling of topics but focuses on improving topics after they have been mined. By employing prompt engineering, we direct LLMs to eliminate off-topic words within a given topic, ensuring that only contextually relevant words are preserved or substituted with ones that fit better semantically. This method emulates human-like scrutiny and improvement of topics, thereby elevating the semantic quality of the topics generated by various models. Our comprehensive evaluation across three unique datasets has shown that our topic refinement approach significantly enhances the semantic coherence of topics.
comment: 6 pages, 4 figures
☆ MEP: Multiple Kernel Learning Enhancing Relative Positional Encoding Length Extrapolation
When the predicted sequence length exceeds the length seen during training, the transformer's inference accuracy diminishes. Existing relative position encoding methods, such as those based on the ALiBi technique, address the length extrapolation challenge exclusively through the implementation of a single kernel function, which introduces a constant bias to every post-softmax attention scores according to their distance. These approaches do not investigate or employ multiple kernel functions to address the extrapolation challenge. Drawing on the ALiBi approach, this study proposes a novel relative positional encoding method, called MEP, which employs a weighted average to combine distinct kernel functions(such as the exponential kernel and the Gaussian kernel) to generate a bias that is applied to post-softmax attention scores. Initially, the framework utilizes various kernel functions to construct multiple kernel functions. Each kernel function adheres to a consistent mean weight coefficient, harnessing the synergistic advantages of different kernels to formulate an innovative bias function. Subsequently, specific slopes are tailored for each kernel function, applying penalties at varying rates, to enhance the model's extrapolation capabilities. Finally, this bias is seamlessly incorporated as a penalty to the post-softmax scores. We present two distinct versions of our method: a parameter-free variant that requires no new learnable parameters, which enhances length extrapolation capabilities without compromising training efficiency, and a parameterized variant capable of integrating state-of-the-art techniques. Empirical evaluations across diverse datasets have demonstrated that both variants of our method achieve state-of-the-art performance, outperforming traditional parameter-free and parameterized approaches.
☆ ExpressEdit: Video Editing with Natural Language and Sketching
Informational videos serve as a crucial source for explaining conceptual and procedural knowledge to novices and experts alike. When producing informational videos, editors edit videos by overlaying text/images or trimming footage to enhance the video quality and make it more engaging. However, video editing can be difficult and time-consuming, especially for novice video editors who often struggle with expressing and implementing their editing ideas. To address this challenge, we first explored how multimodality$-$natural language (NL) and sketching, which are natural modalities humans use for expression$-$can be utilized to support video editors in expressing video editing ideas. We gathered 176 multimodal expressions of editing commands from 10 video editors, which revealed the patterns of use of NL and sketching in describing edit intents. Based on the findings, we present ExpressEdit, a system that enables editing videos via NL text and sketching on the video frame. Powered by LLM and vision models, the system interprets (1) temporal, (2) spatial, and (3) operational references in an NL command and spatial references from sketching. The system implements the interpreted edits, which then the user can iterate on. An observational study (N=10) showed that ExpressEdit enhanced the ability of novice video editors to express and implement their edit ideas. The system allowed participants to perform edits more efficiently and generate more ideas by generating edits based on user's multimodal edit commands and supporting iterations on the editing commands. This work offers insights into the design of future multimodal interfaces and AI-based pipelines for video editing.
comment: 22 pages, 5 figures, to be published in ACM IUI 2024
☆ Solution for Emotion Prediction Competition of Workshop on Emotionally and Culturally Intelligent AI
This report provide a detailed description of the method that we explored and proposed in the WECIA Emotion Prediction Competition (EPC), which predicts a person's emotion through an artistic work with a comment. The dataset of this competition is ArtELingo, designed to encourage work on diversity across languages and cultures. The dataset has two main challenges, namely modal imbalance problem and language-cultural differences problem. In order to address this issue, we propose a simple yet effective approach called single-multi modal with Emotion-Cultural specific prompt(ECSP), which focuses on using the single modal message to enhance the performance of multimodal models and a well-designed prompt to reduce cultural differences problem. To clarify, our approach contains two main blocks: (1)XLM-R\cite{conneau2019unsupervised} based unimodal model and X$^2$-VLM\cite{zeng2022x} based multimodal model (2) Emotion-Cultural specific prompt. Our approach ranked first in the final test with a score of 0.627.
☆ Onboard deep lossless and near-lossless predictive coding of hyperspectral images with line-based attention
Deep learning methods have traditionally been difficult to apply to compression of hyperspectral images onboard of spacecrafts, due to the large computational complexity needed to achieve adequate representational power, as well as the lack of suitable datasets for training and testing. In this paper, we depart from the traditional autoencoder approach and we design a predictive neural network, called LineRWKV, that works recursively line-by-line to limit memory consumption. In order to achieve that, we adopt a novel hybrid attentive-recursive operation that combines the representational advantages of Transformers with the linear complexity and recursive implementation of recurrent neural networks. The compression algorithm performs prediction of each pixel using LineRWKV, followed by entropy coding of the residual. Experiments on the HySpecNet-11k dataset and PRISMA images show that LineRWKV is the first deep-learning method to outperform CCSDS-123.0-B-2 at lossless and near-lossless compression. Promising throughput results are also evaluated on a 7W embedded system.
☆ Depending on yourself when you should: Mentoring LLM with RL agents to become the master in cybersecurity games
Integrating LLM and reinforcement learning (RL) agent effectively to achieve complementary performance is critical in high stake tasks like cybersecurity operations. In this study, we introduce SecurityBot, a LLM agent mentored by pre-trained RL agents, to support cybersecurity operations. In particularly, the LLM agent is supported with a profile module to generated behavior guidelines, a memory module to accumulate local experiences, a reflection module to re-evaluate choices, and an action module to reduce action space. Additionally, it adopts the collaboration mechanism to take suggestions from pre-trained RL agents, including a cursor for dynamic suggestion taken, an aggregator for multiple mentors' suggestions ranking and a caller for proactive suggestion asking. Building on the CybORG experiment framework, our experiences show that SecurityBot demonstrates significant performance improvement compared with LLM or RL standalone, achieving the complementary performance in the cybersecurity games.
comment: 10 pages
☆ Language Models for Text Classification: Is In-Context Learning Enough? LREC
Recent foundational language models have shown state-of-the-art performance in many NLP tasks in zero- and few-shot settings. An advantage of these models over more standard approaches based on fine-tuning is the ability to understand instructions written in natural language (prompts), which helps them generalise better to different tasks and domains without the need for specific training data. This makes them suitable for addressing text classification problems for domains with limited amounts of annotated instances. However, existing research is limited in scale and lacks understanding of how text generation models combined with prompting techniques compare to more established methods for text classification such as fine-tuning masked language models. In this paper, we address this research gap by performing a large-scale evaluation study for 16 text classification datasets covering binary, multiclass, and multilabel problems. In particular, we compare zero- and few-shot approaches of large language models to fine-tuning smaller language models. We also analyse the results by prompt, classification type, domain, and number of labels. In general, the results show how fine-tuning smaller and more efficient language models can still outperform few-shot approaches of larger language models, which have room for improvement when it comes to text classification.
comment: Accepted at LREC-COLING 2024
☆ SGHormer: An Energy-Saving Graph Transformer Driven by Spikes
Graph Transformers (GTs) with powerful representation learning ability make a huge success in wide range of graph tasks. However, the costs behind outstanding performances of GTs are higher energy consumption and computational overhead. The complex structure and quadratic complexity during attention calculation in vanilla transformer seriously hinder its scalability on the large-scale graph data. Though existing methods have made strides in simplifying combinations among blocks or attention-learning paradigm to improve GTs' efficiency, a series of energy-saving solutions originated from biologically plausible structures are rarely taken into consideration when constructing GT framework. To this end, we propose a new spiking-based graph transformer (SGHormer). It turns full-precision embeddings into sparse and binarized spikes to reduce memory and computational costs. The spiking graph self-attention and spiking rectify blocks in SGHormer explicitly capture global structure information and recover the expressive power of spiking embeddings, respectively. In experiments, SGHormer achieves comparable performances to other full-precision GTs with extremely low computational energy consumption. The results show that SGHomer makes a remarkable progress in the field of low-energy GTs.
comment: 9 pages, 3 figures
☆ An Extension-based Approach for Computing and Verifying Preferences in Abstract Argumentation
We present an extension-based approach for computing and verifying preferences in an abstract argumentation system. Although numerous argumentation semantics have been developed previously for identifying acceptable sets of arguments from an argumentation framework, there is a lack of justification behind their acceptability based on implicit argument preferences. Preference-based argumentation frameworks allow one to determine what arguments are justified given a set of preferences. Our research considers the inverse of the standard reasoning problem, i.e., given an abstract argumentation framework and a set of justified arguments, we compute what the possible preferences over arguments are. Furthermore, there is a need to verify (i.e., assess) that the computed preferences would lead to the acceptable sets of arguments. This paper presents a novel approach and algorithm for exhaustively computing and enumerating all possible sets of preferences (restricted to three identified cases) for a conflict-free set of arguments in an abstract argumentation framework. We prove the soundness, completeness and termination of the algorithm. The research establishes that preferences are determined using an extension-based approach after the evaluation phase (acceptability of arguments) rather than stated beforehand. In this work, we focus our research study on grounded, preferred and stable semantics. We show that the complexity of computing sets of preferences is exponential in the number of arguments, and thus, describe an approximate approach and algorithm to compute the preferences. Furthermore, we present novel algorithms for verifying (i.e., assessing) the computed preferences. We provide details of the implementation of the algorithms (source code has been made available), various experiments performed to evaluate the algorithms and the analysis of the results.
☆ S+t-SNE - Bringing dimensionality reduction to data streams
We present S+t-SNE, an adaptation of the t-SNE algorithm designed to handle infinite data streams. The core idea behind S+t-SNE is to update the t-SNE embedding incrementally as new data arrives, ensuring scalability and adaptability to handle streaming scenarios. By selecting the most important points at each step, the algorithm ensures scalability while keeping informative visualisations. Employing a blind method for drift management adjusts the embedding space, facilitating continuous visualisation of evolving data dynamics. Our experimental evaluations demonstrate the effectiveness and efficiency of S+t-SNE. The results highlight its ability to capture patterns in a streaming scenario. We hope our approach offers researchers and practitioners a real-time tool for understanding and interpreting high-dimensional data.
comment: This preprint has not undergone peer review or any post-submission improvements or corrections. We will soon add a link to the final version of this contribution that underwent peer-review and post-acceptance improvements and was presented at IDA2024 (https://ida2024.org/)
☆ PeersimGym: An Environment for Solving the Task Offloading Problem with Reinforcement Learning
Task offloading, crucial for balancing computational loads across devices in networks such as the Internet of Things, poses significant optimization challenges, including minimizing latency and energy usage under strict communication and storage constraints. While traditional optimization falls short in scalability; and heuristic approaches lack in achieving optimal outcomes, Reinforcement Learning (RL) offers a promising avenue by enabling the learning of optimal offloading strategies through iterative interactions. However, the efficacy of RL hinges on access to rich datasets and custom-tailored, realistic training environments. To address this, we introduce PeersimGym, an open-source, customizable simulation environment tailored for developing and optimizing task offloading strategies within computational networks. PeersimGym supports a wide range of network topologies and computational constraints and integrates a \textit{PettingZoo}-based interface for RL agent deployment in both solo and multi-agent setups. Furthermore, we demonstrate the utility of the environment through experiments with Deep Reinforcement Learning agents, showcasing the potential of RL-based approaches to significantly enhance offloading strategies in distributed computing settings. PeersimGym thus bridges the gap between theoretical RL models and their practical applications, paving the way for advancements in efficient task offloading methodologies.
☆ UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps
In this study, we address a gap in existing unsupervised domain adaptation approaches on LiDAR-based 3D object detection, which have predominantly concentrated on adapting between established, high-density autonomous driving datasets. We focus on sparser point clouds, capturing scenarios from different perspectives: not just from vehicles on the road but also from mobile robots on sidewalks, which encounter significantly different environmental conditions and sensor configurations. We introduce Unsupervised Adversarial Domain Adaptation for 3D Object Detection (UADA3D). UADA3D does not depend on pre-trained source models or teacher-student architectures. Instead, it uses an adversarial approach to directly learn domain-invariant features. We demonstrate its efficacy in various adaptation scenarios, showing significant improvements in both self-driving car and mobile robot domains. Our code is open-source and will be available soon.
☆ Data-driven Energy Consumption Modelling for Electric Micromobility using an Open Dataset
The escalating challenges of traffic congestion and environmental degradation underscore the critical importance of embracing E-Mobility solutions in urban spaces. In particular, micro E-Mobility tools such as E-scooters and E-bikes, play a pivotal role in this transition, offering sustainable alternatives for urban commuters. However, the energy consumption patterns for these tools are a critical aspect that impacts their effectiveness in real-world scenarios and is essential for trip planning and boosting user confidence in using these. To this effect, recent studies have utilised physical models customised for specific mobility tools and conditions, but these models struggle with generalization and effectiveness in real-world scenarios due to a notable absence of open datasets for thorough model evaluation and verification. To fill this gap, our work presents an open dataset, collected in Dublin, Ireland, specifically designed for energy modelling research related to E-Scooters and E-Bikes. Furthermore, we provide a comprehensive analysis of energy consumption modelling based on the dataset using a set of representative machine learning algorithms and compare their performance against the contemporary mathematical models as a baseline. Our results demonstrate a notable advantage for data-driven models in comparison to the corresponding mathematical models for estimating energy consumption. Specifically, data-driven models outperform physical models in accuracy by up to 83.83% for E-Bikes and 82.16% for E-Scooters based on an in-depth analysis of the dataset under certain assumptions.
comment: 7 pages, 5 figures, 4 tables. This manuscript has been accepted by the IEEE ITEC 2024
☆ Denoising Table-Text Retrieval for Open-Domain Question Answering LREC
In table-text open-domain question answering, a retriever system retrieves relevant evidence from tables and text to answer questions. Previous studies in table-text open-domain question answering have two common challenges: firstly, their retrievers can be affected by false-positive labels in training datasets; secondly, they may struggle to provide appropriate evidence for questions that require reasoning across the table. To address these issues, we propose Denoised Table-Text Retriever (DoTTeR). Our approach involves utilizing a denoised training dataset with fewer false positive labels by discarding instances with lower question-relevance scores measured through a false positive detection model. Subsequently, we integrate table-level ranking information into the retriever to assist in finding evidence for questions that demand reasoning across the table. To encode this ranking information, we fine-tune a rank-aware column encoder to identify minimum and maximum values within a column. Experimental results demonstrate that DoTTeR significantly outperforms strong baselines on both retrieval recall and downstream QA tasks. Our code is available at https://github.com/deokhk/DoTTeR.
comment: Accepted to LREC-COLING 2024
☆ Fully-fused Multi-Layer Perceptrons on Intel Data Center GPUs
This paper presents a SYCL implementation of Multi-Layer Perceptrons (MLPs), which targets and is optimized for the Intel Data Center GPU Max 1550. To increase the performance, our implementation minimizes the slow global memory accesses by maximizing the data reuse within the general register file and the shared local memory by fusing the operations in each layer of the MLP. We show with a simple roofline model that this results in a significant increase in the arithmetic intensity, leading to improved performance, especially for inference. We compare our approach to a similar CUDA implementation for MLPs and show that our implementation on the Intel Data Center GPU outperforms the CUDA implementation on Nvidia's H100 GPU by a factor up to 2.84 in inference and 1.75 in training. The paper also showcases the efficiency of our SYCL implementation in three significant areas: Image Compression, Neural Radiance Fields, and Physics-Informed Machine Learning. In all cases, our implementation outperforms the off-the-shelf Intel Extension for PyTorch (IPEX) implementation on the same Intel GPU by up to a factor of 30 and the CUDA PyTorch version on Nvidia's H100 GPU by up to a factor 19. The code can be found at https://github.com/intel/tiny-dpcpp-nn.
☆ LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation
Microscopic traffic simulation plays a crucial role in transportation engineering by providing insights into individual vehicle behavior and overall traffic flow. However, creating a realistic simulator that accurately replicates human driving behaviors in various traffic conditions presents significant challenges. Traditional simulators relying on heuristic models often fail to deliver accurate simulations due to the complexity of real-world traffic environments. Due to the covariate shift issue, existing imitation learning-based simulators often fail to generate stable long-term simulations. In this paper, we propose a novel approach called learner-aware supervised imitation learning to address the covariate shift problem in multi-agent imitation learning. By leveraging a variational autoencoder simultaneously modeling the expert and learner state distribution, our approach augments expert states such that the augmented state is aware of learner state distribution. Our method, applied to urban traffic simulation, demonstrates significant improvements over existing state-of-the-art baselines in both short-term microscopic and long-term macroscopic realism when evaluated on the real-world dataset pNEUMA.
comment: accepted by cvpr 2024. arXiv admin note: text overlap with arXiv:2306.06401
☆ Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models CVPR2024
With the emergence of pre-trained vision-language models like CLIP, how to adapt them to various downstream classification tasks has garnered significant attention in recent research. The adaptation strategies can be typically categorized into three paradigms: zero-shot adaptation, few-shot adaptation, and the recently-proposed training-free few-shot adaptation. Most existing approaches are tailored for a specific setting and can only cater to one or two of these paradigms. In this paper, we introduce a versatile adaptation approach that can effectively work under all three settings. Specifically, we propose the dual memory networks that comprise dynamic and static memory components. The static memory caches training data knowledge, enabling training-free few-shot adaptation, while the dynamic memory preserves historical test features online during the testing process, allowing for the exploration of additional data insights beyond the training set. This novel capability enhances model performance in the few-shot setting and enables model usability in the absence of training data. The two memory networks employ the same flexible memory interactive strategy, which can operate in a training-free mode and can be further enhanced by incorporating learnable projection layers. Our approach is tested across 11 datasets under the three task settings. Remarkably, in the zero-shot scenario, it outperforms existing methods by over 3\% and even shows superior results against methods utilizing external training data. Additionally, our method exhibits robust performance against natural distribution shifts. Codes are available at \url{https://github.com/YBZh/DMN}.
comment: CVPR2024; Codes are available at \url{https://github.com/YBZh/DMN}
☆ Parameterized Analysis of Bribery in Challenge the Champ Tournaments
Challenge the champ tournaments are one of the simplest forms of competition, where a (initially selected) champ is repeatedly challenged by other players. If a player beats the champ, then that player is considered the new (current) champ. Each player in the competition challenges the current champ once in a fixed order. The champ of the last round is considered the winner of the tournament. We investigate a setting where players can be bribed to lower their winning probability against the initial champ. The goal is to maximize the probability of the initial champ winning the tournament by bribing the other players, while not exceeding a given budget for the bribes. Mattei et al. [Journal of Applied Logic, 2015] showed that the problem can be solved in pseudo-polynomial time, and that it is in XP when parameterized by the number of players. We show that the problem is weakly NP-hard and W[1]-hard when parameterized by the number of players. On the algorithmic side, we show that the problem is fixed-parameter tractable when parameterized either by the number of different bribe values or the number of different probability values. To this end, we establish several results that are of independent interest. In particular, we show that the product knapsack problem is W[1]-hard when parameterized by the number of items in the knapsack, and that constructive bribery for cup tournaments is W[1]-hard when parameterized by the number of players. Furthermore, we present a novel way of designing mixed integer linear programs, ensuring optimal solutions where all variables are integers.
☆ Towards a Zero-Data, Controllable, Adaptive Dialog System
Conversational Tree Search (V\"ath et al., 2023) is a recent approach to controllable dialog systems, where domain experts shape the behavior of a Reinforcement Learning agent through a dialog tree. The agent learns to efficiently navigate this tree, while adapting to information needs, e.g., domain familiarity, of different users. However, the need for additional training data hinders deployment in new domains. To address this, we explore approaches to generate this data directly from dialog trees. We improve the original approach, and show that agents trained on synthetic data can achieve comparable dialog success to models trained on human data, both when using a commercial Large Language Model for generation, or when using a smaller open-source model, running on a single GPU. We further demonstrate the scalability of our approach by collecting and testing on two new datasets: ONBOARD, a new domain helping foreign residents moving to a new city, and the medical domain DIAGNOSE, a subset of Wikipedia articles related to scalp and head symptoms. Finally, we perform human testing, where no statistically significant differences were found in either objective or subjective measures between models trained on human and generated data.
☆ m3P: Towards Multimodal Multilingual Translation with Multimodal Prompt COLING 2024
Multilingual translation supports multiple translation directions by projecting all languages in a shared space, but the translation quality is undermined by the difference between languages in the text-only modality, especially when the number of languages is large. To bridge this gap, we introduce visual context as the universal language-independent representation to facilitate multilingual translation. In this paper, we propose a framework to leverage the multimodal prompt to guide the Multimodal Multilingual neural Machine Translation (m3P), which aligns the representations of different languages with the same meaning and generates the conditional vision-language memory for translation. We construct a multilingual multimodal instruction dataset (InstrMulti102) to support 102 languages. Our method aims to minimize the representation distance of different languages by regarding the image as a central language. Experimental results show that m3P outperforms previous text-only baselines and multilingual multimodal methods by a large margin. Furthermore, the probing experiments validate the effectiveness of our method in enhancing translation under the low-resource and massively multilingual scenario.
comment: COLING 2024
☆ Practical Applications of Advanced Cloud Services and Generative AI Systems in Medical Image Analysis
The medical field is one of the important fields in the application of artificial intelligence technology. With the explosive growth and diversification of medical data, as well as the continuous improvement of medical needs and challenges, artificial intelligence technology is playing an increasingly important role in the medical field. Artificial intelligence technologies represented by computer vision, natural language processing, and machine learning have been widely penetrated into diverse scenarios such as medical imaging, health management, medical information, and drug research and development, and have become an important driving force for improving the level and quality of medical services.The article explores the transformative potential of generative AI in medical imaging, emphasizing its ability to generate syntheticACM-2 data, enhance images, aid in anomaly detection, and facilitate image-to-image translation. Despite challenges like model complexity, the applications of generative models in healthcare, including Med-PaLM 2 technology, show promising results. By addressing limitations in dataset size and diversity, these models contribute to more accurate diagnoses and improved patient outcomes. However, ethical considerations and collaboration among stakeholders are essential for responsible implementation. Through experiments leveraging GANs to augment brain tumor MRI datasets, the study demonstrates how generative AI can enhance image quality and diversity, ultimately advancing medical diagnostics and patient care.
☆ VDSC: Enhancing Exploration Timing with Value Discrepancy and State Counts
Despite the considerable attention given to the questions of \textit{how much} and \textit{how to} explore in deep reinforcement learning, the investigation into \textit{when} to explore remains relatively less researched. While more sophisticated exploration strategies can excel in specific, often sparse reward environments, existing simpler approaches, such as $\epsilon$-greedy, persist in outperforming them across a broader spectrum of domains. The appeal of these simpler strategies lies in their ease of implementation and generality across a wide range of domains. The downside is that these methods are essentially a blind switching mechanism, which completely disregards the agent's internal state. In this paper, we propose to leverage the agent's internal state to decide \textit{when} to explore, addressing the shortcomings of blind switching mechanisms. We present Value Discrepancy and State Counts through homeostasis (VDSC), a novel approach for efficient exploration timing. Experimental results on the Atari suite demonstrate the superiority of our strategy over traditional methods such as $\epsilon$-greedy and Boltzmann, as well as more sophisticated techniques like Noisy Nets.
☆ KC-GenRe: A Knowledge-constrained Generative Re-ranking Method Based on Large Language Models for Knowledge Graph Completion LREC
The goal of knowledge graph completion (KGC) is to predict missing facts among entities. Previous methods for KGC re-ranking are mostly built on non-generative language models to obtain the probability of each candidate. Recently, generative large language models (LLMs) have shown outstanding performance on several tasks such as information extraction and dialog systems. Leveraging them for KGC re-ranking is beneficial for leveraging the extensive pre-trained knowledge and powerful generative capabilities. However, it may encounter new problems when accomplishing the task, namely mismatch, misordering and omission. To this end, we introduce KC-GenRe, a knowledge-constrained generative re-ranking method based on LLMs for KGC. To overcome the mismatch issue, we formulate the KGC re-ranking task as a candidate identifier sorting generation problem implemented by generative LLMs. To tackle the misordering issue, we develop a knowledge-guided interactive training method that enhances the identification and ranking of candidates. To address the omission issue, we design a knowledge-augmented constrained inference method that enables contextual prompting and controlled generation, so as to obtain valid rankings. Experimental results show that KG-GenRe achieves state-of-the-art performance on four datasets, with gains of up to 6.7% and 7.7% in the MRR and Hits@1 metric compared to previous methods, and 9.0% and 11.1% compared to that without re-ranking. Extensive analysis demonstrates the effectiveness of components in KG-GenRe.
comment: This paper has been accepted for publication in the proceedings of LREC-COLING 2024
☆ Boosting Few-Shot Learning with Disentangled Self-Supervised Learning and Meta-Learning for Medical Image Classification
Background and objective: Employing deep learning models in critical domains such as medical imaging poses challenges associated with the limited availability of training data. We present a strategy for improving the performance and generalization capabilities of models trained in low-data regimes. Methods: The proposed method starts with a pre-training phase, where features learned in a self-supervised learning setting are disentangled to improve the robustness of the representations for downstream tasks. We then introduce a meta-fine-tuning step, leveraging related classes between meta-training and meta-testing phases but varying the granularity level. This approach aims to enhance the model's generalization capabilities by exposing it to more challenging classification tasks during meta-training and evaluating it on easier tasks but holding greater clinical relevance during meta-testing. We demonstrate the effectiveness of the proposed approach through a series of experiments exploring several backbones, as well as diverse pre-training and fine-tuning schemes, on two distinct medical tasks, i.e., classification of prostate cancer aggressiveness from MRI data and classification of breast cancer malignity from microscopic images. Results: Our results indicate that the proposed approach consistently yields superior performance w.r.t. ablation experiments, maintaining competitiveness even when a distribution shift between training and evaluation data occurs. Conclusion: Extensive experiments demonstrate the effectiveness and wide applicability of the proposed approach. We hope that this work will add another solution to the arsenal of addressing learning issues in data-scarce imaging domains.
comment: 20 pages, 4 figures, 4 tables. Submitted to Elsevier on 25 March 2024
☆ Equipping Sketch Patches with Context-Aware Positional Encoding for Graphic Sketch Representation
The drawing order of a sketch records how it is created stroke-by-stroke by a human being. For graphic sketch representation learning, recent studies have injected sketch drawing orders into graph edge construction by linking each patch to another in accordance to a temporal-based nearest neighboring strategy. However, such constructed graph edges may be unreliable, since a sketch could have variants of drawings. In this paper, we propose a variant-drawing-protected method by equipping sketch patches with context-aware positional encoding (PE) to make better use of drawing orders for learning graphic sketch representation. Instead of injecting sketch drawings into graph edges, we embed these sequential information into graph nodes only. More specifically, each patch embedding is equipped with a sinusoidal absolute PE to highlight the sequential position in the drawing order. And its neighboring patches, ranked by the values of self-attention scores between patch embeddings, are equipped with learnable relative PEs to restore the contextual positions within a neighborhood. During message aggregation via graph convolutional networks, a node receives both semantic contents from patch embeddings and contextual patterns from PEs by its neighbors, arriving at drawing-order-enhanced sketch representations. Experimental results indicate that our method significantly improves sketch healing and controllable sketch synthesis.
☆ MapGuide: A Simple yet Effective Method to Reconstruct Continuous Language from Brain Activities NAACL 2024
Decoding continuous language from brain activity is a formidable yet promising field of research. It is particularly significant for aiding people with speech disabilities to communicate through brain signals. This field addresses the complex task of mapping brain signals to text. The previous best attempt reverse-engineered this process in an indirect way: it began by learning to encode brain activity from text and then guided text generation by aligning with predicted brain responses. In contrast, we propose a simple yet effective method that guides text reconstruction by directly comparing them with the predicted text embeddings mapped from brain activities. Comprehensive experiments reveal that our method significantly outperforms the current state-of-the-art model, showing average improvements of 77% and 54% on BLEU and METEOR scores. We further validate the proposed modules through detailed ablation studies and case analyses and highlight a critical correlation: the more precisely we map brain activities to text embeddings, the better the text reconstruction results. Such insight can simplify the task of reconstructing language from brain activities for future work, emphasizing the importance of improving brain-to-text-embedding mapping techniques.
comment: Accepted to NAACL 2024 main conference
☆ Prediction-sharing During Training and Inference
Two firms are engaged in a competitive prediction task. Each firm has two sources of data -- labeled historical data and unlabeled inference-time data -- and uses the former to derive a prediction model, and the latter to make predictions on new instances. We study data-sharing contracts between the firms. The novelty of our study is to introduce and highlight the differences between contracts that share prediction models only, contracts to share inference-time predictions only, and contracts to share both. Our analysis proceeds on three levels. First, we develop a general Bayesian framework that facilitates our study. Second, we narrow our focus to two natural settings within this framework: (i) a setting in which the accuracy of each firm's prediction model is common knowledge, but the correlation between the respective models is unknown; and (ii) a setting in which two hypotheses exist regarding the optimal predictor, and one of the firms has a structural advantage in deducing it. Within these two settings we study optimal contract choice. More specifically, we find the individually rational and Pareto-optimal contracts for some notable cases, and describe specific settings where each of the different sharing contracts emerge as optimal. Finally, in the third level of our analysis we demonstrate the applicability of our concepts in a synthetic simulation using real loan data.
☆ Natural Language Requirements Testability Measurement Based on Requirement Smells
Requirements form the basis for defining software systems' obligations and tasks. Testable requirements help prevent failures, reduce maintenance costs, and make it easier to perform acceptance tests. However, despite the importance of measuring and quantifying requirements testability, no automatic approach for measuring requirements testability has been proposed based on the requirements smells, which are at odds with the requirements testability. This paper presents a mathematical model to evaluate and rank the natural language requirements testability based on an extensive set of nine requirements smells, detected automatically, and acceptance test efforts determined by requirement length and its application domain. Most of the smells stem from uncountable adjectives, context-sensitive, and ambiguous words. A comprehensive dictionary is required to detect such words. We offer a neural word-embedding technique to generate such a dictionary automatically. Using the dictionary, we could automatically detect Polysemy smell (domain-specific ambiguity) for the first time in 10 application domains. Our empirical study on nearly 1000 software requirements from six well-known industrial and academic projects demonstrates that the proposed smell detection approach outperforms Smella, a state-of-the-art tool, in detecting requirements smells. The precision and recall of smell detection are improved with an average of 0.03 and 0.33, respectively, compared to the state-of-the-art. The proposed requirement testability model measures the testability of 985 requirements with a mean absolute error of 0.12 and a mean squared error of 0.03, demonstrating the model's potential for practical use.
comment: 45 pages, 16 figures, and 13 tables; submitted as a journal paper
☆ A Unified Kernel for Neural Network Learning
Past decades have witnessed a great interest in the distinction and connection between neural network learning and kernel learning. Recent advancements have made theoretical progress in connecting infinite-wide neural networks and Gaussian processes. Two predominant approaches have emerged: the Neural Network Gaussian Process (NNGP) and the Neural Tangent Kernel (NTK). The former, rooted in Bayesian inference, represents a zero-order kernel, while the latter, grounded in the tangent space of gradient descents, is a first-order kernel. In this paper, we present the Unified Neural Kernel (UNK), which characterizes the learning dynamics of neural networks with gradient descents and parameter initialization. The proposed UNK kernel maintains the limiting properties of both NNGP and NTK, exhibiting behaviors akin to NTK with a finite learning step and converging to NNGP as the learning step approaches infinity. Besides, we also theoretically characterize the uniform tightness and learning convergence of the UNK kernel, providing comprehensive insights into this unified kernel. Experimental results underscore the effectiveness of our proposed method.
☆ LaRE^2: Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection CVPR 2024
The evolution of Diffusion Models has dramatically improved image generation quality, making it increasingly difficult to differentiate between real and generated images. This development, while impressive, also raises significant privacy and security concerns. In response to this, we propose a novel Latent REconstruction error guided feature REfinement method (LaRE^2) for detecting the diffusion-generated images. We come up with the Latent Reconstruction Error (LaRE), the first reconstruction-error based feature in the latent space for generated image detection. LaRE surpasses existing methods in terms of feature extraction efficiency while preserving crucial cues required to differentiate between the real and the fake. To exploit LaRE, we propose an Error-Guided feature REfinement module (EGRE), which can refine the image feature guided by LaRE to enhance the discriminativeness of the feature. Our EGRE utilizes an align-then-refine mechanism, which effectively refines the image feature for generated-image detection from both spatial and channel perspectives. Extensive experiments on the large-scale GenImage benchmark demonstrate the superiority of our LaRE^2, which surpasses the best SoTA method by up to 11.9%/12.1% average ACC/AP across 8 different image generators. LaRE also surpasses existing methods in terms of feature extraction cost, delivering an impressive speed enhancement of 8 times.
comment: CVPR 2024
☆ Imitating Cost-Constrained Behaviors in Reinforcement Learning ICAPS-24
Complex planning and scheduling problems have long been solved using various optimization or heuristic approaches. In recent years, imitation learning that aims to learn from expert demonstrations has been proposed as a viable alternative to solving these problems. Generally speaking, imitation learning is designed to learn either the reward (or preference) model or directly the behavioral policy by observing the behavior of an expert. Existing work in imitation learning and inverse reinforcement learning has focused on imitation primarily in unconstrained settings (e.g., no limit on fuel consumed by the vehicle). However, in many real-world domains, the behavior of an expert is governed not only by reward (or preference) but also by constraints. For instance, decisions on self-driving delivery vehicles are dependent not only on the route preferences/rewards (depending on past demand data) but also on the fuel in the vehicle and the time available. In such problems, imitation learning is challenging as decisions are not only dictated by the reward model but are also dependent on a cost-constrained model. In this paper, we provide multiple methods that match expert distributions in the presence of trajectory cost constraints through (a) Lagrangian-based method; (b) Meta-gradients to find a good trade-off between expected return and minimizing constraint violation; and (c) Cost-violation-based alternating gradient. We empirically show that leading imitation learning approaches imitate cost-constrained behaviors poorly and our meta-gradient-based approach achieves the best performance.
comment: Accepted to the 34th International Conference on Automated Planning and Scheduling (ICAPS-24)
☆ Incorporating Exponential Smoothing into MLP: A Simple but Effective Sequence Model
Modeling long-range dependencies in sequential data is a crucial step in sequence learning. A recently developed model, the Structured State Space (S4), demonstrated significant effectiveness in modeling long-range sequences. However, It is unclear whether the success of S4 can be attributed to its intricate parameterization and HiPPO initialization or simply due to State Space Models (SSMs). To further investigate the potential of the deep SSMs, we start with exponential smoothing (ETS), a simple SSM, and propose a stacked architecture by directly incorporating it into an element-wise MLP. We augment simple ETS with additional parameters and complex field to reduce the inductive bias. Despite increasing less than 1\% of parameters of element-wise MLP, our models achieve comparable results to S4 on the LRA benchmark.
comment: 12 pages, 5 tables, 3 figures
☆ Aligning Large Language Models for Enhancing Psychiatric Interviews through Symptom Delineation and Summarization
Recent advancements in Large Language Models (LLMs) have accelerated their usage in various domains. Given the fact that psychiatric interviews are goal-oriented and structured dialogues between the professional interviewer and the interviewee, it is one of the most underexplored areas where LLMs can contribute substantial value. Here, we explore the use of LLMs for enhancing psychiatric interviews, by analyzing counseling data from North Korean defectors with traumatic events and mental health issues. Specifically, we investigate whether LLMs can (1) delineate the part of the conversation that suggests psychiatric symptoms and name the symptoms, and (2) summarize stressors and symptoms, based on the interview dialogue transcript. Here, the transcript data was labeled by mental health experts for training and evaluation of LLMs. Our experimental results show that appropriately prompted LLMs can achieve high performance on both the symptom delineation task and the summarization task. This research contributes to the nascent field of applying LLMs to psychiatric interview and demonstrates their potential effectiveness in aiding mental health practitioners.
☆ Knowledge-Powered Recommendation for an Improved Diet Water Footprint AAAI'24
According to WWF, 1.1 billion people lack access to water, and 2.7 billion experience water scarcity at least one month a year. By 2025, two-thirds of the world's population may be facing water shortages. This highlights the urgency of managing water usage efficiently, especially in water-intensive sectors like food. This paper proposes a recommendation engine, powered by knowledge graphs, aiming to facilitate sustainable and healthy food consumption. The engine recommends ingredient substitutes in user recipes that improve nutritional value and reduce environmental impact, particularly water footprint. The system architecture includes source identification, information extraction, schema alignment, knowledge graph construction, and user interface development. The research offers a promising tool for promoting healthier eating habits and contributing to water conservation efforts.
comment: 3 pages, 1 figure, AAAI'24
☆ MA4DIV: Multi-Agent Reinforcement Learning for Search Result Diversification
The objective of search result diversification (SRD) is to ensure that selected documents cover as many different subtopics as possible. Existing methods primarily utilize a paradigm of "greedy selection", i.e., selecting one document with the highest diversity score at a time. These approaches tend to be inefficient and are easily trapped in a suboptimal state. In addition, some other methods aim to approximately optimize the diversity metric, such as $\alpha$-NDCG, but the results still remain suboptimal. To address these challenges, we introduce Multi-Agent reinforcement learning (MARL) for search result DIVersity, which called MA4DIV. In this approach, each document is an agent and the search result diversification is modeled as a cooperative task among multiple agents. This approach allows for directly optimizing the diversity metrics, such as $\alpha$-NDCG, while achieving high training efficiency. We conducted preliminary experiments on public TREC datasets to demonstrate the effectiveness and potential of MA4DIV. Considering the limited number of queries in public TREC datasets, we construct a large-scale dataset from industry sources and show that MA4DIV achieves substantial improvements in both effectiveness and efficiency than existing baselines on a industrial scale dataset.
☆ AI Safety: Necessary, but insufficient and possibly problematic
This article critically examines the recent hype around AI safety. We first start with noting the nature of the AI safety hype as being dominated by governments and corporations, and contrast it with other avenues within AI research on advancing social good. We consider what 'AI safety' actually means, and outline the dominant concepts that the digital footprint of AI safety aligns with. We posit that AI safety has a nuanced and uneasy relationship with transparency and other allied notions associated with societal good, indicating that it is an insufficient notion if the goal is that of societal good in a broad sense. We note that the AI safety debate has already influenced some regulatory efforts in AI, perhaps in not so desirable directions. We also share our concerns on how AI safety may normalize AI that advances structural harm through providing exploitative and harmful AI with a veneer of safety.
☆ On permutation-invariant neural networks
Conventional machine learning algorithms have traditionally been designed under the assumption that input data follows a vector-based format, with an emphasis on vector-centric paradigms. However, as the demand for tasks involving set-based inputs has grown, there has been a paradigm shift in the research community towards addressing these challenges. In recent years, the emergence of neural network architectures such as Deep Sets and Transformers has presented a significant advancement in the treatment of set-based data. These architectures are specifically engineered to naturally accommodate sets as input, enabling more effective representation and processing of set structures. Consequently, there has been a surge of research endeavors dedicated to exploring and harnessing the capabilities of these architectures for various tasks involving the approximation of set functions. This comprehensive survey aims to provide an overview of the diverse problem settings and ongoing research efforts pertaining to neural networks that approximate set functions. By delving into the intricacies of these approaches and elucidating the associated challenges, the survey aims to equip readers with a comprehensive understanding of the field. Through this comprehensive perspective, we hope that researchers can gain valuable insights into the potential applications, inherent limitations, and future directions of set-based neural networks. Indeed, from this survey we gain two insights: i) Deep Sets and its variants can be generalized by differences in the aggregation function, and ii) the behavior of Deep Sets is sensitive to the choice of the aggregation function. From these observations, we show that Deep Sets, one of the well-known permutation-invariant neural networks, can be generalized in the sense of a quasi-arithmetic mean.
☆ Transcribing Bengali Text with Regional Dialects to IPA using District Guided Tokens
Accurate transcription of Bengali text to the International Phonetic Alphabet (IPA) is a challenging task due to the complex phonology of the language and context-dependent sound changes. This challenge is even more for regional Bengali dialects due to unavailability of standardized spelling conventions for these dialects, presence of local and foreign words popular in those regions and phonological diversity across different regions. This paper presents an approach to this sequence-to-sequence problem by introducing the District Guided Tokens (DGT) technique on a new dataset spanning six districts of Bangladesh. The key idea is to provide the model with explicit information about the regional dialect or "district" of the input text before generating the IPA transcription. This is achieved by prepending a district token to the input sequence, effectively guiding the model to understand the unique phonetic patterns associated with each district. The DGT technique is applied to fine-tune several transformer-based models, on this new dataset. Experimental results demonstrate the effectiveness of DGT, with the ByT5 model achieving superior performance over word-based models like mT5, BanglaT5, and umT5. This is attributed to ByT5's ability to handle a high percentage of out-of-vocabulary words in the test set. The proposed approach highlights the importance of incorporating regional dialect information into ubiquitous natural language processing systems for languages with diverse phonological variations. The following work was a result of the "Bhashamul" challenge, which is dedicated to solving the problem of Bengali text with regional dialects to IPA transcription https://www.kaggle.com/competitions/regipa/. The training and inference notebooks are available through the competition link.
comment: This work became the champion of the Bhashamul challenge
☆ An Open-source End-to-End Logic Optimization Framework for Large-scale Boolean Network with Reinforcement Learning
We propose an open-source end-to-end logic optimization framework for large-scale boolean network with reinforcement learning.
comment: 5 pages, 4 figures, 1 table
☆ ELLEN: Extremely Lightly Supervised Learning For Efficient Named Entity Recognition LREC
In this work, we revisit the problem of semi-supervised named entity recognition (NER) focusing on extremely light supervision, consisting of a lexicon containing only 10 examples per class. We introduce ELLEN, a simple, fully modular, neuro-symbolic method that blends fine-tuned language models with linguistic rules. These rules include insights such as ''One Sense Per Discourse'', using a Masked Language Model as an unsupervised NER, leveraging part-of-speech tags to identify and eliminate unlabeled entities as false negatives, and other intuitions about classifier confidence scores in local and global context. ELLEN achieves very strong performance on the CoNLL-2003 dataset when using the minimal supervision from the lexicon above. It also outperforms most existing (and considerably more complex) semi-supervised NER methods under the same supervision settings commonly used in the literature (i.e., 5% of the training data). Further, we evaluate our CoNLL-2003 model in a zero-shot scenario on WNUT-17 where we find that it outperforms GPT-3.5 and achieves comparable performance to GPT-4. In a zero-shot setting, ELLEN also achieves over 75% of the performance of a strong, fully supervised model trained on gold data. Our code is available at: https://github.com/hriaz17/ELLEN.
comment: Accepted to LREC-COLING 2024
☆ Explainable Graph Neural Networks for Observation Impact Analysis in Atmospheric State Estimation
This paper investigates the impact of observations on atmospheric state estimation in weather forecasting systems using graph neural networks (GNNs) and explainability methods. We integrate observation and Numerical Weather Prediction (NWP) points into a meteorological graph, extracting $k$-hop subgraphs centered on NWP points. Self-supervised GNNs are employed to estimate the atmospheric state by aggregating data within these $k$-hop radii. The study applies gradient-based explainability methods to quantify the significance of different observations in the estimation process. Evaluated with data from 11 satellite and land-based observations, the results highlight the effectiveness of visualizing the importance of observation types, enhancing the understanding and optimization of observational data in weather forecasting.
☆ Application-Driven Innovation in Machine Learning
As applications of machine learning proliferate, innovative algorithms inspired by specific real-world challenges have become increasingly important. Such work offers the potential for significant impact not merely in domains of application but also in machine learning itself. In this paper, we describe the paradigm of application-driven research in machine learning, contrasting it with the more standard paradigm of methods-driven research. We illustrate the benefits of application-driven machine learning and how this approach can productively synergize with methods-driven work. Despite these benefits, we find that reviewing, hiring, and teaching practices in machine learning often hold back application-driven innovation. We outline how these processes may be improved.
comment: 12 pages, 3 figures
☆ Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance
Recent studies have demonstrated that diffusion models are capable of generating high-quality samples, but their quality heavily depends on sampling guidance techniques, such as classifier guidance (CG) and classifier-free guidance (CFG). These techniques are often not applicable in unconditional generation or in various downstream tasks such as image restoration. In this paper, we propose a novel sampling guidance, called Perturbed-Attention Guidance (PAG), which improves diffusion sample quality across both unconditional and conditional settings, achieving this without requiring additional training or the integration of external modules. PAG is designed to progressively enhance the structure of samples throughout the denoising process. It involves generating intermediate samples with degraded structure by substituting selected self-attention maps in diffusion U-Net with an identity matrix, by considering the self-attention mechanisms' ability to capture structural information, and guiding the denoising process away from these degraded samples. In both ADM and Stable Diffusion, PAG surprisingly improves sample quality in conditional and even unconditional scenarios. Moreover, PAG significantly improves the baseline performance in various downstream tasks where existing guidances such as CG or CFG cannot be fully utilized, including ControlNet with empty prompts and image restoration such as inpainting and deblurring.
comment: Project page is available at https://ku-cvlab.github.io/Perturbed-Attention-Guidance
☆ AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving CVPR-2024
Autonomous vehicle (AV) systems rely on robust perception models as a cornerstone of safety assurance. However, objects encountered on the road exhibit a long-tailed distribution, with rare or unseen categories posing challenges to a deployed perception model. This necessitates an expensive process of continuously curating and annotating data with significant human effort. We propose to leverage recent advances in vision-language and large language models to design an Automatic Data Engine (AIDE) that automatically identifies issues, efficiently curates data, improves the model through auto-labeling, and verifies the model through generation of diverse scenarios. This process operates iteratively, allowing for continuous self-improvement of the model. We further establish a benchmark for open-world detection on AV datasets to comprehensively evaluate various learning paradigms, demonstrating our method's superior performance at a reduced cost.
comment: Accepted by CVPR-2024
☆ ChatGPT Rates Natural Language Explanation Quality Like Humans: But on Which Scales? LREC
As AI becomes more integral in our lives, the need for transparency and responsibility grows. While natural language explanations (NLEs) are vital for clarifying the reasoning behind AI decisions, evaluating them through human judgments is complex and resource-intensive due to subjectivity and the need for fine-grained ratings. This study explores the alignment between ChatGPT and human assessments across multiple scales (i.e., binary, ternary, and 7-Likert scale). We sample 300 data instances from three NLE datasets and collect 900 human annotations for both informativeness and clarity scores as the text quality measurement. We further conduct paired comparison experiments under different ranges of subjectivity scores, where the baseline comes from 8,346 human annotations. Our results show that ChatGPT aligns better with humans in more coarse-grained scales. Also, paired comparisons and dynamic prompting (i.e., providing semantically similar examples in the prompt) improve the alignment. This research advances our understanding of large language models' capabilities to assess the text explanation quality in different configurations for responsible AI development.
comment: Accpeted by LREC-COLING 2024 main conference, long paper
☆ Bridging Textual and Tabular Worlds for Fact Verification: A Lightweight, Attention-Based Model LREC
FEVEROUS is a benchmark and research initiative focused on fact extraction and verification tasks involving unstructured text and structured tabular data. In FEVEROUS, existing works often rely on extensive preprocessing and utilize rule-based transformations of data, leading to potential context loss or misleading encodings. This paper introduces a simple yet powerful model that nullifies the need for modality conversion, thereby preserving the original evidence's context. By leveraging pre-trained models on diverse text and tabular datasets and by incorporating a lightweight attention-based mechanism, our approach efficiently exploits latent connections between different data types, thereby yielding comprehensive and reliable verdict predictions. The model's modular structure adeptly manages multi-modal information, ensuring the integrity and authenticity of the original evidence are uncompromised. Comparative analyses reveal that our approach exhibits competitive performance, aligning itself closely with top-tier models on the FEVEROUS benchmark.
comment: Accepted for a presentation at LREC-COLING 2024 - The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation
☆ Addressing Myopic Constrained POMDP Planning with Recursive Dual Ascent ICAPS
Lagrangian-guided Monte Carlo tree search with global dual ascent has been applied to solve large constrained partially observable Markov decision processes (CPOMDPs) online. In this work, we demonstrate that these global dual parameters can lead to myopic action selection during exploration, ultimately leading to suboptimal decision making. To address this, we introduce history-dependent dual variables that guide local action selection and are optimized with recursive dual ascent. We empirically compare the performance of our approach on a motivating toy example and two large CPOMDPs, demonstrating improved exploration, and ultimately, safer outcomes.
comment: Accepted to the 2024 International Conference on Automated Planning and Scheduling (ICAPS)
☆ MESIA: Understanding and Leveraging Supplementary Nature of Method-level Comments for Automatic Comment Generation
Code comments are important for developers in program comprehension. In scenarios of comprehending and reusing a method, developers expect code comments to provide supplementary information beyond the method signature. However, the extent of such supplementary information varies a lot in different code comments. In this paper, we raise the awareness of the supplementary nature of method-level comments and propose a new metric named MESIA (Mean Supplementary Information Amount) to assess the extent of supplementary information that a code comment can provide. With the MESIA metric, we conduct experiments on a popular code-comment dataset and three common types of neural approaches to generate method-level comments. Our experimental results demonstrate the value of our proposed work with a number of findings. (1) Small-MESIA comments occupy around 20% of the dataset and mostly fall into only the WHAT comment category. (2) Being able to provide various kinds of essential information, large-MESIA comments in the dataset are difficult for existing neural approaches to generate. (3) We can improve the capability of existing neural approaches to generate large-MESIA comments by reducing the proportion of small-MESIA comments in the training set. (4) The retrained model can generate large-MESIA comments that convey essential meaningful supplementary information for methods in the small-MESIA test set, but will get a lower BLEU score in evaluation. These findings indicate that with good training data, auto-generated comments can sometimes even surpass human-written reference comments, and having no appropriate ground truth for evaluation is an issue that needs to be addressed by future work on automatic comment generation.
comment: In 32nd IEEE/ACM International Conference on Program Comprehension (ICPC'24)
☆ The Solution of the Zodiac Killer's 340-Character Cipher
The case of the Zodiac Killer is one of the most widely known unsolved serial killer cases in history. The unidentified killer murdered five known victims and terrorized the state of California. He also communicated extensively with the press and law enforcement. Besides his murders, Zodiac was known for his use of ciphers. The first Zodiac cipher was solved within a week of its publication, while the second cipher was solved by the authors after 51 years, when it was discovered to be a transposition and homophonic substitution cipher with unusual qualities. In this paper, we detail the historical significance of this cipher and the numerous efforts which culminated in its solution.
☆ The Solution for the ICCV 2023 1st Scientific Figure Captioning Challenge
In this paper, we propose a solution for improving the quality of captions generated for figures in papers. We adopt the approach of summarizing the textual content in the paper to generate image captions. Throughout our study, we encounter discrepancies in the OCR information provided in the official dataset. To rectify this, we employ the PaddleOCR toolkit to extract OCR information from all images. Moreover, we observe that certain textual content in the official paper pertains to images that are not relevant for captioning, thereby introducing noise during caption generation. To mitigate this issue, we leverage LLaMA to extract image-specific information by querying the textual content based on image mentions, effectively filtering out extraneous information. Additionally, we recognize a discrepancy between the primary use of maximum likelihood estimation during text generation and the evaluation metrics such as ROUGE employed to assess the quality of generated captions. To bridge this gap, we integrate the BRIO model framework, enabling a more coherent alignment between the generation and evaluation processes. Our approach ranked first in the final test with a score of 4.49.
Reinforcement Learning-based Receding Horizon Control using Adaptive Control Barrier Functions for Safety-Critical Systems
Optimal control methods provide solutions to safety-critical problems but easily become intractable. Control Barrier Functions (CBFs) have emerged as a popular technique that facilitates their solution by provably guaranteeing safety, through their forward invariance property, at the expense of some performance loss. This approach involves defining a performance objective alongside CBF-based safety constraints that must always be enforced. Unfortunately, both performance and solution feasibility can be significantly impacted by two key factors: (i) the selection of the cost function and associated parameters, and (ii) the calibration of parameters within the CBF-based constraints, which capture the trade-off between performance and conservativeness. %as well as infeasibility. To address these challenges, we propose a Reinforcement Learning (RL)-based Receding Horizon Control (RHC) approach leveraging Model Predictive Control (MPC) with CBFs (MPC-CBF). In particular, we parameterize our controller and use bilevel optimization, where RL is used to learn the optimal parameters while MPC computes the optimal control input. We validate our method by applying it to the challenging automated merging control problem for Connected and Automated Vehicles (CAVs) at conflicting roadways. Results demonstrate improved performance and a significant reduction in the number of infeasible cases compared to traditional heuristic approaches used for tuning CBF-based controllers, showcasing the effectiveness of the proposed method.
☆ The Pursuit of Fairness in Artificial Intelligence Models: A Survey
Artificial Intelligence (AI) models are now being utilized in all facets of our lives such as healthcare, education and employment. Since they are used in numerous sensitive environments and make decisions that can be life altering, potential biased outcomes are a pressing matter. Developers should ensure that such models don't manifest any unexpected discriminatory practices like partiality for certain genders, ethnicities or disabled people. With the ubiquitous dissemination of AI systems, researchers and practitioners are becoming more aware of unfair models and are bound to mitigate bias in them. Significant research has been conducted in addressing such issues to ensure models don't intentionally or unintentionally perpetuate bias. This survey offers a synopsis of the different ways researchers have promoted fairness in AI systems. We explore the different definitions of fairness existing in the current literature. We create a comprehensive taxonomy by categorizing different types of bias and investigate cases of biased AI in different application domains. A thorough study is conducted of the approaches and techniques employed by researchers to mitigate bias in AI models. Moreover, we also delve into the impact of biased models on user experience and the ethical considerations to contemplate when developing and deploying such models. We hope this survey helps researchers and practitioners understand the intricate details of fairness and bias in AI systems. By sharing this thorough survey, we aim to promote additional discourse in the domain of equitable and responsible AI.
comment: 37 pages, 6 figures
☆ Deep Support Vectors
While the success of deep learning is commonly attributed to its theoretical equivalence with Support Vector Machines (SVM), the practical implications of this relationship have not been thoroughly explored. This paper pioneers an exploration in this domain, specifically focusing on the identification of Deep Support Vectors (DSVs) within deep learning models. We introduce the concept of DeepKKT conditions, an adaptation of the traditional Karush-Kuhn-Tucker (KKT) conditions tailored for deep learning. Through empirical investigations, we illustrate that DSVs exhibit similarities to support vectors in SVM, offering a tangible method to interpret the decision-making criteria of models. Additionally, our findings demonstrate that models can be effectively reconstructed using DSVs, resembling the process in SVM. The code will be available.
☆ Learning Traffic Signal Control via Genetic Programming
The control of traffic signals is crucial for improving transportation efficiency. Recently, learning-based methods, especially Deep Reinforcement Learning (DRL), garnered substantial success in the quest for more efficient traffic signal control strategies. However, the design of rewards in DRL highly demands domain knowledge to converge to an effective policy, and the final policy also presents difficulties in terms of explainability. In this work, a new learning-based method for signal control in complex intersections is proposed. In our approach, we design a concept of phase urgency for each signal phase. During signal transitions, the traffic light control strategy selects the next phase to be activated based on the phase urgency. We then proposed to represent the urgency function as an explainable tree structure. The urgency function can calculate the phase urgency for a specific phase based on the current road conditions. Genetic programming is adopted to perform gradient-free optimization of the urgency function. We test our algorithm on multiple public traffic signal control datasets. The experimental results indicate that the tree-shaped urgency function evolved by genetic programming outperforms the baselines, including a state-of-the-art method in the transportation field and a well-known DRL-based method.
☆ JMultiWOZ: A Large-Scale Japanese Multi-Domain Task-Oriented Dialogue Dataset LREC
Dialogue datasets are crucial for deep learning-based task-oriented dialogue system research. While numerous English language multi-domain task-oriented dialogue datasets have been developed and contributed to significant advancements in task-oriented dialogue systems, such a dataset does not exist in Japanese, and research in this area is limited compared to that in English. In this study, towards the advancement of research and development of task-oriented dialogue systems in Japanese, we constructed JMultiWOZ, the first Japanese language large-scale multi-domain task-oriented dialogue dataset. Using JMultiWOZ, we evaluated the dialogue state tracking and response generation capabilities of the state-of-the-art methods on the existing major English benchmark dataset MultiWOZ2.2 and the latest large language model (LLM)-based methods. Our evaluation results demonstrated that JMultiWOZ provides a benchmark that is on par with MultiWOZ2.2. In addition, through evaluation experiments of interactive dialogues with the models and human participants, we identified limitations in the task completion capabilities of LLMs in Japanese.
comment: Accepted by LREC-COLING 2024
☆ Oh! We Freeze: Improving Quantized Knowledge Distillation via Signal Propagation Analysis for Large Language Models ICLR 2024
Large generative models, such as large language models (LLMs) and diffusion models have as revolutionized the fields of NLP and computer vision respectively. However, their slow inference, high computation and memory requirement makes it challenging to deploy them on edge devices. In this study, we propose a light-weight quantization aware fine tuning technique using knowledge distillation (KD-QAT) to improve the performance of 4-bit weight quantized LLMs using commonly available datasets to realize a popular language use case, on device chat applications. To improve this paradigm of finetuning, as main contributions, we provide insights into stability of KD-QAT by empirically studying the gradient propagation during training to better understand the vulnerabilities of KD-QAT based approaches to low-bit quantization errors. Based on our insights, we propose ov-freeze, a simple technique to stabilize the KD-QAT process. Finally, we experiment with the popular 7B LLaMAv2-Chat model at 4-bit quantization level and demonstrate that ov-freeze results in near float-point precision performance, i.e., less than 0.7% loss of accuracy on Commonsense Reasoning benchmarks.
comment: Accepted at Practical ML for Low Resource Settings Workshop at ICLR 2024
☆ Large Language Models Produce Responses Perceived to be Empathic
Large Language Models (LLMs) have demonstrated surprising performance on many tasks, including writing supportive messages that display empathy. Here, we had these models generate empathic messages in response to posts describing common life experiences, such as workplace situations, parenting, relationships, and other anxiety- and anger-eliciting situations. Across two studies (N=192, 202), we showed human raters a variety of responses written by several models (GPT4 Turbo, Llama2, and Mistral), and had people rate these responses on how empathic they seemed to be. We found that LLM-generated responses were consistently rated as more empathic than human-written responses. Linguistic analyses also show that these models write in distinct, predictable ``styles", in terms of their use of punctuation, emojis, and certain words. These results highlight the potential of using LLMs to enhance human peer support in contexts where empathy is important.
☆ A Real-Time Rescheduling Algorithm for Multi-robot Plan Execution ICAPS 2024
One area of research in multi-agent path finding is to determine how replanning can be efficiently achieved in the case of agents being delayed during execution. One option is to reschedule the passing order of agents, i.e., the sequence in which agents visit the same location. In response, we propose Switchable-Edge Search (SES), an A*-style algorithm designed to find optimal passing orders. We prove the optimality of SES and evaluate its efficiency via simulations. The best variant of SES takes less than 1 second for small- and medium-sized problems and runs up to 4 times faster than baselines for large-sized problems.
comment: ICAPS 2024
☆ Juru: Legal Brazilian Large Language Model from Reputable Sources
The high computational cost associated with pretraining large language models limits their research. Two strategies have emerged to address this issue: domain specialization and pretraining with high-quality data. To explore these strategies, we specialized the Sabi\'a-2 Small model with 1.9 billion unique tokens from reputable Brazilian legal sources and conducted few-shot evaluations on legal and general knowledge exams. Our model, Juru, demonstrates the benefits of domain specialization with a reduced amount of pretraining data. However, this specialization comes at the expense of degrading performance in other knowledge areas within the same language. This study contributes to the growing body of scientific evidence showing that pretraining data selection may enhance the performance of large language models, enabling the exploration of these models at a lower cost.
☆ Securing GNNs: Explanation-Based Identification of Backdoored Training Graphs
Graph Neural Networks (GNNs) have gained popularity in numerous domains, yet they are vulnerable to backdoor attacks that can compromise their performance and ethical application. The detection of these attacks is crucial for maintaining the reliability and security of GNN classification tasks, but effective detection techniques are lacking. Following an initial investigation, we observed that while graph-level explanations can offer limited insights, their effectiveness in detecting backdoor triggers is inconsistent and incomplete. To bridge this gap, we extract and transform secondary outputs of GNN explanation mechanisms, designing seven novel metrics that more effectively detect backdoor attacks. Additionally, we develop an adaptive attack to rigorously evaluate our approach. We test our method on multiple benchmark datasets and examine its efficacy against various attack models. Our results show that our method can achieve high detection performance, marking a significant advancement in safeguarding GNNs against backdoor attacks.
☆ AE SemRL: Learning Semantic Association Rules with Autoencoders
Association Rule Mining (ARM) is the task of learning associations among data features in the form of logical rules. Mining association rules from high-dimensional numerical data, for example, time series data from a large number of sensors in a smart environment, is a computationally intensive task. In this study, we propose an Autoencoder-based approach to learn and extract association rules from time series data (AE SemRL). Moreover, we argue that in the presence of semantic information related to time series data sources, semantics can facilitate learning generalizable and explainable association rules. Despite enriching time series data with additional semantic features, AE SemRL makes learning association rules from high-dimensional data feasible. Our experiments show that semantic association rules can be extracted from a latent representation created by an Autoencoder and this method has in the order of hundreds of times faster execution time than state-of-the-art ARM approaches in many scenarios. We believe that this study advances a new way of extracting associations from representations and has the potential to inspire more research in this field.
☆ Recommendation of data-free class-incremental learning algorithms by simulating future data
Class-incremental learning deals with sequential data streams composed of batches of classes. Various algorithms have been proposed to address the challenging case where samples from past classes cannot be stored. However, selecting an appropriate algorithm for a user-defined setting is an open problem, as the relative performance of these algorithms depends on the incremental settings. To solve this problem, we introduce an algorithm recommendation method that simulates the future data stream. Given an initial set of classes, it leverages generative models to simulate future classes from the same visual domain. We evaluate recent algorithms on the simulated stream and recommend the one which performs best in the user-defined incremental setting. We illustrate the effectiveness of our method on three large datasets using six algorithms and six incremental settings. Our method outperforms competitive baselines, and performance is close to that of an oracle choosing the best algorithm in each setting. This work contributes to facilitate the practical deployment of incremental learning.
☆ Don't Trust: Verify -- Grounding LLM Quantitative Reasoning with Autoformalization ICLR 2024
Large language models (LLM), such as Google's Minerva and OpenAI's GPT families, are becoming increasingly capable of solving mathematical quantitative reasoning problems. However, they still make unjustified logical and computational errors in their reasoning steps and answers. In this paper, we leverage the fact that if the training corpus of LLMs contained sufficiently many examples of formal mathematics (e.g. in Isabelle, a formal theorem proving environment), they can be prompted to translate i.e. autoformalize informal mathematical statements into formal Isabelle code -- which can be verified automatically for internal consistency. This provides a mechanism to automatically reject solutions whose formalized versions are inconsistent within themselves or with the formalized problem statement. We evaluate our method on GSM8K, MATH and MultiArith datasets and demonstrate that our approach provides a consistently better heuristic than vanilla majority voting -- the previously best method to identify correct answers, by more than 12% on GSM8K. In our experiments it improves results consistently across all datasets and LLM model sizes. The code can be found at https://github.com/jinpz/dtv.
comment: ICLR 2024
♻ ☆ Simple and Scalable Strategies to Continually Pre-train Large Language Models
Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant compute compared to re-training. However, the distribution shift induced by new data typically results in degraded performance on previous data or poor adaptation to the new data. In this work, we show that a simple and scalable combination of learning rate (LR) re-warming, LR re-decaying, and replay of previous data is sufficient to match the performance of fully re-training from scratch on all available data, as measured by the final loss and the average score on several language model (LM) evaluation benchmarks. Specifically, we show this for a weak but realistic distribution shift between two commonly used LLM pre-training datasets (English$\rightarrow$English) and a stronger distribution shift (English$\rightarrow$German) at the $405$M parameter model scale with large dataset sizes (hundreds of billions of tokens). Selecting the weak but realistic shift for larger-scale experiments, we also find that our continual learning strategies match the re-training baseline for a 10B parameter LLM. Our results demonstrate that LLMs can be successfully updated via simple and scalable continual learning strategies, matching the re-training baseline using only a fraction of the compute. Finally, inspired by previous work, we propose alternatives to the cosine learning rate schedule that help circumvent forgetting induced by LR re-warming and that are not bound to a fixed token budget.
♻ ☆ DiVa-360: The Dynamic Visual Dataset for Immersive Neural Fields
Advances in neural fields are enabling high-fidelity capture of the shape and appearance of dynamic 3D scenes. However, their capabilities lag behind those offered by conventional representations such as 2D videos because of algorithmic challenges and the lack of large-scale multi-view real-world datasets. We address the dataset limitation with DiVa-360, a real-world 360 dynamic visual dataset that contains synchronized high-resolution and long-duration multi-view video sequences of table-scale scenes captured using a customized low-cost system with 53 cameras. It contains 21 object-centric sequences categorized by different motion types, 25 intricate hand-object interaction sequences, and 8 long-duration sequences for a total of 17.4 M image frames. In addition, we provide foreground-background segmentation masks, synchronized audio, and text descriptions. We benchmark the state-of-the-art dynamic neural field methods on DiVa-360 and provide insights about existing methods and future challenges on long-duration neural field capture.
♻ ☆ Dynamics of Moral Behavior in Heterogeneous Populations of Learning Agents
Growing concerns about safety and alignment of AI systems highlight the importance of embedding moral capabilities in artificial agents. A promising solution is the use of learning from experience, i.e., Reinforcement Learning. In multi-agent (social) environments, complex population-level phenomena may emerge from interactions between individual learning agents. Many of the existing studies rely on simulated social dilemma environments to study the interactions of independent learning agents. However, they tend to ignore the moral heterogeneity that is likely to be present in societies of agents in practice. For example, at different points in time a single learning agent may face opponents who are consequentialist (i.e., caring about maximizing some outcome over time) or norm-based (i.e., focusing on conforming to a specific norm here and now). The extent to which agents' co-development may be impacted by such moral heterogeneity in populations is not well understood. In this paper, we present a study of the learning dynamics of morally heterogeneous populations interacting in a social dilemma setting. Using a Prisoner's Dilemma environment with a partner selection mechanism, we investigate the extent to which the prevalence of diverse moral agents in populations affects individual agents' learning behaviors and emergent population-level outcomes. We observe several types of non-trivial interactions between pro-social and anti-social agents, and find that certain classes of moral agents are able to steer selfish agents towards more cooperative behavior.
♻ ☆ Safe Explicable Planning
Human expectations arise from their understanding of others and the world. In the context of human-AI interaction, this understanding may not align with reality, leading to the AI agent failing to meet expectations and compromising team performance. Explicable planning, introduced as a method to bridge this gap, aims to reconcile human expectations with the agent's optimal behavior, facilitating interpretable decision-making. However, an unresolved critical issue is ensuring safety in explicable planning, as it could result in explicable behaviors that are unsafe. To address this, we propose Safe Explicable Planning (SEP), which extends the prior work to support the specification of a safety bound. The goal of SEP is to find behaviors that align with human expectations while adhering to the specified safety criterion. Our approach generalizes the consideration of multiple objectives stemming from multiple models rather than a single model, yielding a Pareto set of safe explicable policies. We present both an exact method, guaranteeing finding the Pareto set, and a more efficient greedy method that finds one of the policies in the Pareto set. Additionally, we offer approximate solutions based on state aggregation to improve scalability. We provide formal proofs that validate the desired theoretical properties of these methods. Evaluation through simulations and physical robot experiments confirms the effectiveness of our approach for safe explicable planning.
♻ ☆ AI and Generative AI for Research Discovery and Summarization
AI and generative AI tools, including chatbots like ChatGPT that rely on large language models (LLMs), have burst onto the scene this year, creating incredible opportunities to increase work productivity and improve our lives. Statisticians and data scientists have begun experiencing the benefits from the availability of these tools in numerous ways, such as the generation of programming code from text prompts to analyze data or fit statistical models. One area that these tools can make a substantial impact is in research discovery and summarization. Standalone tools and plugins to chatbots are being developed that allow researchers to more quickly find relevant literature than pre-2023 search tools. Furthermore, generative AI tools have improved to the point where they can summarize and extract the key points from research articles in succinct language. Finally, chatbots based on highly parameterized LLMs can be used to simulate abductive reasoning, which provides researchers the ability to make connections among related technical topics, which can also be used for research discovery. We review the developments in AI and generative AI for research discovery and summarization, and propose directions where these types of tools are likely to head in the future that may be of interest to statistician and data scientists.
comment: 29 pages, 9 figures
♻ ☆ ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review
ChatGPT is another large language model (LLM) vastly available for the consumers on their devices but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail the issues and concerns raised over chatGPT in line with aforementioned characteristics. We also discuss the recent EU AI Act briefly in accordance with the SPADE evaluation. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for AI policy act, if designed by the governments.
comment: 29 pages, 8 figures, 4 tables
♻ ☆ Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics
DeepFakes, which refer to AI-generated media content, have become an increasing concern due to their use as a means for disinformation. Detecting DeepFakes is currently solved with programmed machine learning algorithms. In this work, we investigate the capabilities of multimodal large language models (LLMs) in DeepFake detection. We conducted qualitative and quantitative experiments to demonstrate multimodal LLMs and show that they can expose AI-generated images through careful experimental design and prompt engineering. This is interesting, considering that LLMs are not inherently tailored for media forensic tasks, and the process does not require programming. We discuss the limitations of multimodal LLMs for these tasks and suggest possible improvements.
♻ ☆ Efficient Pre-training for Localized Instruction Generation of Videos
Procedural videos show step-by-step demonstrations of tasks like recipe preparation. Understanding such videos is challenging, involving the precise localization of steps and the generation of textual instructions. Manually annotating steps and writing instructions is costly, which limits the size of current datasets and hinders effective learning. Leveraging large but noisy video-transcript datasets for pre-training can boost performance, but demands significant computational resources. Furthermore, transcripts contain irrelevant content and exhibit style variation compared to instructions written by human annotators. To mitigate both issues, we propose a technique, Sieve-&-Swap, to automatically curate a smaller dataset: (i) Sieve filters irrelevant transcripts and (ii) Swap enhances the quality of the text instruction by automatically replacing the transcripts with human-written instructions from a text-only recipe dataset. The curated dataset, three orders of magnitude smaller than current web-scale datasets, enables efficient training of large-scale models with competitive performance. We complement our Sieve-\&-Swap approach with a Procedure Transformer (ProcX) for end-to-end step localization and instruction generation for procedural videos. When this model is pre-trained on our curated dataset, it achieves state-of-the-art performance in zero-shot and finetuning settings on YouCook2 and Tasty, while using a fraction of the computational resources.
comment: This version has some missing experiments and elaborative technical details
♻ ☆ Analyzing the Quality Attributes of AI Vision Models in Open Repositories Under Adversarial Attacks
As AI models rapidly evolve, they are frequently released to open repositories, such as HuggingFace. It is essential to perform quality assurance validation on these models before integrating them into the production development lifecycle. In addition to evaluating efficiency in terms of balanced accuracy and computing costs, adversarial attacks are potential threats to the robustness and explainability of AI models. Meanwhile, XAI applies algorithms that approximate inputs to outputs post-hoc to identify the contributing features. Adversarial perturbations may also degrade the utility of XAI explanations that require further investigation. In this paper, we present an integrated process designed for downstream evaluation tasks, including validating AI model accuracy, evaluating robustness with benchmark perturbations, comparing explanation utility, and assessing overhead. We demonstrate an evaluation scenario involving six computer vision models, which include CNN-based, Transformer-based, and hybrid architectures, three types of perturbations, and five XAI methods, resulting in ninety unique combinations. The process reveals the explanation utility among the XAI methods in terms of the identified key areas responding to the adversarial perturbation. The process produces aggregated results that illustrate multiple attributes of each AI model.
♻ ☆ Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts for Open-Domain QA?
While auxiliary information has become a key to enhancing Large Language Models (LLMs), relatively little is known about how LLMs merge these contexts, specifically contexts generated by LLMs and those retrieved from external sources. To investigate this, we formulate a systematic framework to identify whether LLMs' responses, derived from the integration of generated and retrieved contexts, are attributed to either generated or retrieved contexts. To easily trace the origin of the response, we construct datasets with conflicting contexts, i.e., each question is paired with both generated and retrieved contexts, yet only one of them contains the correct answer. Our experiments reveal a significant bias in several LLMs (GPT-4/3.5 and Llama2) to favor generated contexts, even when they provide incorrect information. We further identify two key factors contributing to this bias: i) contexts generated by LLMs typically show greater similarity to the questions, increasing their likelihood of being selected; ii) the segmentation process used in retrieved contexts disrupts their completeness, thereby hindering their full utilization in LLMs. Our analysis enhances the understanding of how LLMs merge diverse contexts, offering valuable insights for advancing current augmentation methods for LLMs.
♻ ☆ Toward a Theory of Causation for Interpreting Neural Code Models
Neural Language Models of Code, or Neural Code Models (NCMs), are rapidly progressing from research prototypes to commercial developer tools. As such, understanding the capabilities and limitations of such models is becoming critical. However, the abilities of these models are typically measured using automated metrics that often only reveal a portion of their real-world performance. While, in general, the performance of NCMs appears promising, currently much is unknown about how such models arrive at decisions. To this end, this paper introduces $do_{code}$, a post hoc interpretability method specific to NCMs that is capable of explaining model predictions. $do_{code}$ is based upon causal inference to enable programming language-oriented explanations. While the theoretical underpinnings of $do_{code}$ are extensible to exploring different model properties, we provide a concrete instantiation that aims to mitigate the impact of spurious correlations by grounding explanations of model behavior in properties of programming languages. To demonstrate the practical benefit of $do_{code}$, we illustrate the insights that our framework can provide by performing a case study on two popular deep learning architectures and ten NCMs. The results of this case study illustrate that our studied NCMs are sensitive to changes in code syntax. All our NCMs, except for the BERT-like model, statistically learn to predict tokens related to blocks of code (\eg brackets, parenthesis, semicolon) with less confounding bias as compared to other programming language constructs. These insights demonstrate the potential of $do_{code}$ as a useful method to detect and facilitate the elimination of confounding bias in NCMs.
comment: Accepted to appear in IEEE Transactions on Software Engineering
♻ ☆ Cyber-Security Knowledge Graph Generation by Hierarchical Nonnegative Matrix Factorization
Much of human knowledge in cybersecurity is encapsulated within the ever-growing volume of scientific papers. As this textual data continues to expand, the importance of document organization methods becomes increasingly crucial for extracting actionable insights hidden within large text datasets. Knowledge Graphs (KGs) serve as a means to store factual information in a structured manner, providing explicit, interpretable knowledge that includes domain-specific information from the cybersecurity scientific literature. One of the challenges in constructing a KG from scientific literature is the extraction of ontology from unstructured text. In this paper, we address this topic and introduce a method for building a multi-modal KG by extracting structured ontology from scientific papers. We demonstrate this concept in the cybersecurity domain. One modality of the KG represents observable information from the papers, such as the categories in which they were published or the authors. The second modality uncovers latent (hidden) patterns of text extracted through hierarchical and semantic non-negative matrix factorization (NMF), such as named entities, topics or clusters, and keywords. We illustrate this concept by consolidating more than two million scientific papers uploaded to arXiv into the cyber-domain, using hierarchical and semantic NMF, and by building a cyber-domain-specific KG.
comment: Accepted at IEEE ISDFS
♻ ☆ Decode Neural signal as Speech
Decoding language from brain dynamics is an important open direction in the realm of brain-computer interface (BCI), especially considering the rapid growth of large language models. Compared to invasive-based signals which require electrode implantation surgery, non-invasive neural signals (e.g. EEG, MEG) have attracted increasing attention considering their safety and generality. However, the exploration is not adequate in three aspects: 1) previous methods mainly focus on EEG but none of the previous works address this problem on MEG with better signal quality; 2) prior works have predominantly used ``teacher-forcing" during generative decoding, which is impractical; 3) prior works are mostly ``BART-based" not fully auto-regressive, which performs better in other sequence tasks. In this paper, we explore the brain-to-text translation of MEG signals in a speech-decoding formation. Here we are the first to investigate a cross-attention-based ``whisper" model for generating text directly from MEG signals without teacher forcing. Our model achieves impressive BLEU-1 scores of 60.30 and 52.89 without pretraining \& teacher-forcing on two major datasets (\textit{GWilliams} and \textit{Schoffelen}). This paper conducts a comprehensive review to understand how speech decoding formation performs on the neural decoding tasks, including pretraining initialization, training \& evaluation set splitting, augmentation, and scaling law.
♻ ☆ As Good As A Coin Toss: Human detection of AI-generated images, videos, audio, and audiovisual stimuli
As synthetic media becomes progressively more realistic and barriers to using it continue to lower, the technology has been increasingly utilized for malicious purposes, from financial fraud to nonconsensual pornography. Today, the principal defense against being misled by synthetic media relies on the ability of the human observer to visually and auditorily discern between real and fake. However, it remains unclear just how vulnerable people actually are to deceptive synthetic media in the course of their day to day lives. We conducted a perceptual study with 1276 participants to assess how accurate people were at distinguishing synthetic images, audio only, video only, and audiovisual stimuli from authentic. To reflect the circumstances under which people would likely encounter synthetic media in the wild, testing conditions and stimuli emulated a typical online platform, while all synthetic media used in the survey was sourced from publicly accessible generative AI technology. We find that overall, participants struggled to meaningfully discern between synthetic and authentic content. We also find that detection performance worsens when the stimuli contains synthetic content as compared to authentic content, images featuring human faces as compared to non face objects, a single modality as compared to multimodal stimuli, mixed authenticity as compared to being fully synthetic for audiovisual stimuli, and features foreign languages as compared to languages the observer is fluent in. Finally, we also find that prior knowledge of synthetic media does not meaningfully impact their detection performance. Collectively, these results indicate that people are highly susceptible to being tricked by synthetic media in their daily lives and that human perceptual detection capabilities can no longer be relied upon as an effective counterdefense.
comment: For study pre-registration, see https://osf.io/fnhr3
♻ ☆ RL$^3$: Boosting Meta Reinforcement Learning via RL inside RL$^2$
Meta reinforcement learning (meta-RL) methods such as RL$^2$ have emerged as promising approaches for learning data-efficient RL algorithms tailored to a given task distribution. However, they show poor asymptotic performance and struggle with out-of-distribution tasks because they rely on sequence models, such as recurrent neural networks or transformers, to process experiences rather than summarize them using general-purpose RL components such as value functions. In contrast, traditional RL algorithms are data-inefficient as they do not use domain knowledge, but they do converge to an optimal policy in the limit. We propose RL$^3$, a principled hybrid approach that incorporates action-values, learned per task through traditional RL, in the inputs to meta-RL. We show that RL$^3$ earns greater cumulative reward in the long term, compared to RL$^2$, while maintaining data-efficiency in the short term, and generalizes better to out-of-distribution tasks. Experiments are conducted on both custom and benchmark discrete domains from the meta-RL literature that exhibit a range of short-term, long-term, and complex dependencies.
♻ ☆ Multi-Objective Optimization for Sparse Deep Multi-Task Learning
Different conflicting optimization criteria arise naturally in various Deep Learning scenarios. These can address different main tasks (i.e., in the setting of Multi-Task Learning), but also main and secondary tasks such as loss minimization versus sparsity. The usual approach is a simple weighting of the criteria, which formally only works in the convex setting. In this paper, we present a Multi-Objective Optimization algorithm using a modified Weighted Chebyshev scalarization for training Deep Neural Networks (DNNs) with respect to several tasks. By employing this scalarization technique, the algorithm can identify all optimal solutions of the original problem while reducing its complexity to a sequence of single-objective problems. The simplified problems are then solved using an Augmented Lagrangian method, enabling the use of popular optimization techniques such as Adam and Stochastic Gradient Descent, while efficaciously handling constraints. Our work aims to address the (economical and also ecological) sustainability issue of DNN models, with a particular focus on Deep Multi-Task models, which are typically designed with a very large number of weights to perform equally well on multiple tasks. Through experiments conducted on two Machine Learning datasets, we demonstrate the possibility of adaptively sparsifying the model during training without significantly impacting its performance, if we are willing to apply task-specific adaptations to the network weights. Code is available at https://github.com/salomonhotegni/MDMTN
comment: 12 pages, 7 figures
♻ ☆ MedPromptX: Grounded Multimodal Prompting for Chest X-ray Diagnosis
Chest X-ray images are commonly used for predicting acute and chronic cardiopulmonary conditions, but efforts to integrate them with structured clinical data face challenges due to incomplete electronic health records (EHR). This paper introduces \textbf{MedPromptX}, the first model to integrate multimodal large language models (MLLMs), few-shot prompting (FP) and visual grounding (VG) to combine imagery with EHR data for chest X-ray diagnosis. A pre-trained MLLM is utilized to complement the missing EHR information, providing a comprehensive understanding of patients' medical history. Additionally, FP reduces the necessity for extensive training of MLLMs while effectively tackling the issue of hallucination. Nevertheless, the process of determining the optimal number of few-shot examples and selecting high-quality candidates can be burdensome, yet it profoundly influences model performance. Hence, we propose a new technique that dynamically refines few-shot data for real-time adjustment to new patient scenarios. Moreover, VG aids in focusing the model's attention on relevant regions of interest in X-ray images, enhancing the identification of abnormalities. We release MedPromptX-VQA, a new in-context visual question answering dataset encompassing interleaved image and EHR data derived from MIMIC-IV and MIMIC-CXR databases. Results demonstrate the SOTA performance of MedPromptX, achieving an 11% improvement in F1-score compared to the baselines. Code and data are available at https://github.com/BioMedIA-MBZUAI/MedPromptX
♻ ☆ Accelerating Scientific Discovery with Generative Knowledge Extraction, Graph-Based Representation, and Multimodal Intelligent Graph Reasoning
Leveraging generative Artificial Intelligence (AI), we have transformed a dataset comprising 1,000 scientific papers into an ontological knowledge graph. Through an in-depth structural analysis, we have calculated node degrees, identified communities and connectivities, and evaluated clustering coefficients and betweenness centrality of pivotal nodes, uncovering fascinating knowledge architectures. The graph has an inherently scale-free nature, is highly connected, and can be used for graph reasoning by taking advantage of transitive and isomorphic properties that reveal unprecedented interdisciplinary relationships that can be used to answer queries, identify gaps in knowledge, propose never-before-seen material designs, and predict material behaviors. We compute deep node embeddings for combinatorial node similarity ranking for use in a path sampling strategy links dissimilar concepts that have previously not been related. One comparison revealed structural parallels between biological materials and Beethoven's 9th Symphony, highlighting shared patterns of complexity through isomorphic mapping. In another example, the algorithm proposed a hierarchical mycelium-based composite based on integrating path sampling with principles extracted from Kandinsky's 'Composition VII' painting. The resulting material integrates an innovative set of concepts that include a balance of chaos/order, adjustable porosity, mechanical strength, and complex patterned chemical functionalization. We uncover other isomorphisms across science, technology and art, revealing a nuanced ontology of immanence that reveal a context-dependent heterarchical interplay of constituents. Graph-based generative AI achieves a far higher degree of novelty, explorative capacity, and technical detail, than conventional approaches and establishes a widely useful framework for innovation by revealing hidden connections.
♻ ☆ Text-Guided Variational Image Generation for Industrial Anomaly Detection and Segmentation CVPR 2024
We propose a text-guided variational image generation method to address the challenge of getting clean data for anomaly detection in industrial manufacturing. Our method utilizes text information about the target object, learned from extensive text library documents, to generate non-defective data images resembling the input image. The proposed framework ensures that the generated non-defective images align with anticipated distributions derived from textual and image-based knowledge, ensuring stability and generality. Experimental results demonstrate the effectiveness of our approach, surpassing previous methods even with limited non-defective data. Our approach is validated through generalization tests across four baseline models and three distinct datasets. We present an additional analysis to enhance the effectiveness of anomaly detection models by utilizing the generated images.
comment: 18 pages, Accepted to CVPR 2024
♻ ☆ Unveiling the Pitfalls of Knowledge Editing for Large Language Models ICLR 2024
As the cost associated with fine-tuning Large Language Models (LLMs) continues to rise, recent research efforts have pivoted towards developing methodologies to edit implicit knowledge embedded within LLMs. Yet, there's still a dark cloud lingering overhead -- will knowledge editing trigger butterfly effect? since it is still unclear whether knowledge editing might introduce side effects that pose potential risks or not. This paper pioneers the investigation into the potential pitfalls associated with knowledge editing for LLMs. To achieve this, we introduce new benchmark datasets and propose innovative evaluation metrics. Our results underline two pivotal concerns: (1) Knowledge Conflict: Editing groups of facts that logically clash can magnify the inherent inconsistencies in LLMs-a facet neglected by previous methods. (2) Knowledge Distortion: Altering parameters with the aim of editing factual knowledge can irrevocably warp the innate knowledge structure of LLMs. Experimental results vividly demonstrate that knowledge editing might inadvertently cast a shadow of unintended consequences on LLMs, which warrant attention and efforts for future works. Code and data are available at https://github.com/zjunlp/PitfallsKnowledgeEditing.
comment: ICLR 2024
♻ ☆ Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration NAACL 2024
Human intelligence thrives on cognitive synergy, where collaboration among different minds yield superior outcomes compared to isolated individuals. In this work, we propose Solo Performance Prompting (SPP), which transforms a single LLM into a cognitive synergist by engaging in multi-turn self-collaboration with multiple personas. A cognitive synergist is an intelligent agent that collaboratively combines multiple minds' strengths and knowledge to enhance problem-solving in complex tasks. By dynamically identifying and simulating different personas based on task inputs, SPP unleashes the potential of cognitive synergy in LLMs. Our in-depth analysis shows that assigning multiple fine-grained personas in LLMs improves problem-solving abilities compared to using a single or fixed number of personas. We evaluate SPP on three challenging tasks: Trivia Creative Writing, Codenames Collaborative, and Logic Grid Puzzle, encompassing both knowledge-intensive and reasoning-intensive types. Unlike previous works, such as Chain-of-Thought, that solely enhance the reasoning abilities in LLMs, experimental results demonstrate that SPP effectively reduces factual hallucination, and maintains strong reasoning capabilities. Additionally, comparative experiments show that cognitive synergy only emerges in GPT-4 and does not appear in less capable models, such as GPT-3.5-turbo and Llama2-13b-chat, which draws an interesting analogy to human development. Code, data, and prompts can be found at: https://github.com/MikeWangWZHL/Solo-Performance-Prompting.git.
comment: Accepted as a main conference paper at NAACL 2024
♻ ☆ A Safe Preference Learning Approach for Personalization with Applications to Autonomous Vehicles
This work introduces a preference learning method that ensures adherence to given specifications, with an application to autonomous vehicles. Our approach incorporates the priority ordering of Signal Temporal Logic (STL) formulas describing traffic rules into a learning framework. By leveraging Parametric Weighted Signal Temporal Logic (PWSTL), we formulate the problem of safety-guaranteed preference learning based on pairwise comparisons and propose an approach to solve this learning problem. Our approach finds a feasible valuation for the weights of the given PWSTL formula such that, with these weights, preferred signals have weighted quantitative satisfaction measures greater than their non-preferred counterparts. The feasible valuation of weights given by our approach leads to a weighted STL formula that can be used in correct-and-custom-by-construction controller synthesis. We demonstrate the performance of our method with a pilot human subject study in two different simulated driving scenarios involving a stop sign and a pedestrian crossing. Our approach yields competitive results compared to existing preference learning methods in terms of capturing preferences and notably outperforms them when safety is considered.
comment: 9 pages, 3 figures, 2 tables. This work has been published at IEEE Robotics and Automation Letters. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Artificial Neural Nets and the Representation of Human Concepts
What do artificial neural networks (ANNs) learn? The machine learning (ML) community shares the narrative that ANNs must develop abstract human concepts to perform complex tasks. Some go even further and believe that these concepts are stored in individual units of the network. Based on current research, I systematically investigate the assumptions underlying this narrative. I conclude that ANNs are indeed capable of performing complex prediction tasks, and that they may learn human and non-human concepts to do so. However, evidence indicates that ANNs do not represent these concepts in individual units.
comment: For: Philosophy of Science for Machine Learning: Core Issues and New Perspectives, edited by Juan Duran and Giorgia Pozzi
♻ ☆ ViT-Lens: Towards Omni-modal Representations CVPR2024
Aiming to advance AI agents, large foundation models significantly improve reasoning and instruction execution, yet the current focus on vision and language neglects the potential of perceiving diverse modalities in open-world environments. However, the success of data-driven vision and language models is costly or even infeasible to be reproduced for rare modalities. In this paper, we present ViT-Lens-2 that facilitates efficient omni-modal representation learning by perceiving novel modalities with a pretrained ViT and aligning them to a pre-defined space. Specifically, the modality-specific lens is tuned to project any-modal signals to an intermediate embedding space, which are then processed by a strong ViT with pre-trained visual knowledge. The encoded representations are optimized toward aligning with the modal-independent space, pre-defined by off-the-shelf foundation models. ViT-Lens-2 provides a unified solution for representation learning of increasing modalities with two appealing advantages: (i) Unlocking the great potential of pretrained ViTs to novel modalities effectively with efficient data regime; (ii) Enabling emergent downstream capabilities through modality alignment and shared ViT parameters. We tailor ViT-Lens-2 to learn representations for 3D point cloud, depth, audio, tactile and EEG, and set new state-of-the-art results across various understanding tasks, such as zero-shot classification. By seamlessly integrating ViT-Lens-2 into Multimodal Foundation Models, we enable Any-modality to Text and Image Generation in a zero-shot manner. Code and models are available at https://github.com/TencentARC/ViT-Lens.
comment: This work is a follow-up of arXiv:2308.10185. Accepted to CVPR2024
♻ ☆ AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation CVPR 2024
This paper proposes a novel direct Audio-Visual Speech to Audio-Visual Speech Translation (AV2AV) framework, where the input and output of the system are multimodal (i.e., audio and visual speech). With the proposed AV2AV, two key advantages can be brought: 1) We can perform real-like conversations with individuals worldwide in a virtual meeting by utilizing our own primary languages. In contrast to Speech-to-Speech Translation (A2A), which solely translates between audio modalities, the proposed AV2AV directly translates between audio-visual speech. This capability enhances the dialogue experience by presenting synchronized lip movements along with the translated speech. 2) We can improve the robustness of the spoken language translation system. By employing the complementary information of audio-visual speech, the system can effectively translate spoken language even in the presence of acoustic noise, showcasing robust performance. To mitigate the problem of the absence of a parallel AV2AV translation dataset, we propose to train our spoken language translation system with the audio-only dataset of A2A. This is done by learning unified audio-visual speech representations through self-supervised learning in advance to train the translation system. Moreover, we propose an AV-Renderer that can generate raw audio and video in parallel. It is designed with zero-shot speaker modeling, thus the speaker in source audio-visual speech can be maintained at the target translated audio-visual speech. The effectiveness of AV2AV is evaluated with extensive experiments in a many-to-many language translation setting. Demo page is available on https://choijeongsoo.github.io/av2av.
comment: CVPR 2024. Code & Demo: https://choijeongsoo.github.io/av2av
♻ ☆ Coarse-Tuning for Ad-hoc Document Retrieval Using Pre-trained Language Models LREC
Fine-tuning in information retrieval systems using pre-trained language models (PLM-based IR) requires learning query representations and query-document relations, in addition to downstream task-specific learning. This study introduces coarse-tuning as an intermediate learning stage that bridges pre-training and fine-tuning. By learning query representations and query-document relations in coarse-tuning, we aim to reduce the load of fine-tuning and improve the learning effect of downstream IR tasks. We propose Query-Document Pair Prediction (QDPP) for coarse-tuning, which predicts the appropriateness of query-document pairs. Evaluation experiments show that the proposed method significantly improves MRR and/or nDCG@5 in four ad-hoc document retrieval datasets. Furthermore, the results of the query prediction task suggested that coarse-tuning facilitated learning of query representation and query-document relations.
comment: Accepted at LREC-COLING 2024
♻ ☆ Towards more Practical Threat Models in Artificial Intelligence Security
Recent works have identified a gap between research and practice in artificial intelligence security: threats studied in academia do not always reflect the practical use and security risks of AI. For example, while models are often studied in isolation, they form part of larger ML pipelines in practice. Recent works also brought forward that adversarial manipulations introduced by academic attacks are impractical. We take a first step towards describing the full extent of this disparity. To this end, we revisit the threat models of the six most studied attacks in AI security research and match them to AI usage in practice via a survey with 271 industrial practitioners. On the one hand, we find that all existing threat models are indeed applicable. On the other hand, there are significant mismatches: research is often too generous with the attacker, assuming access to information not frequently available in real-world settings. Our paper is thus a call for action to study more practical threat models in artificial intelligence security.
comment: 18 pages, 4 figures, 8 tables, accepted to Usenix Security, incorporated external feedback
♻ ☆ TagAlign: Improving Vision-Language Alignment with Multi-Tag Classification
The crux of learning vision-language models is to extract semantically aligned information from visual and linguistic data. Existing attempts usually face the problem of coarse alignment, e.g., the vision encoder struggles in localizing an attribute-specified object. In this work, we propose an embarrassingly simple approach to better align image and text features with no need of additional data formats other than image-text pairs. Concretely, given an image and its paired text, we manage to parse objects (e.g., cat) and attributes (e.g., black) from the description, which are highly likely to exist in the image. It is noteworthy that the parsing pipeline is fully automatic and thus enjoys good scalability. With these parsed semantics as supervision signals, we can complement the commonly used image-text contrastive loss with the multi-tag classification loss. Extensive experimental results on a broad suite of semantic segmentation datasets substantiate the average 5.2\% improvement of our framework over existing alternatives. Furthermore, the visualization results indicate that attribute supervision makes vision-language models accurately localize attribute-specified objects. Project page can be found at https://qinying-liu.github.io/Tag-Align.
♻ ☆ SGS-SLAM: Semantic Gaussian Splatting For Neural Dense SLAM
We present SGS-SLAM, the first semantic visual SLAM system based on Gaussian Splatting. It incorporates appearance, geometry, and semantic features through multi-channel optimization, addressing the oversmoothing limitations of neural implicit SLAM systems in high-quality rendering, scene understanding, and object-level geometry. We introduce a unique semantic feature loss that effectively compensates for the shortcomings of traditional depth and color losses in object optimization. Through a semantic-guided keyframe selection strategy, we prevent erroneous reconstructions caused by cumulative errors. Extensive experiments demonstrate that SGS-SLAM delivers state-of-the-art performance in camera pose estimation, map reconstruction, precise semantic segmentation, and object-level geometric accuracy, while ensuring real-time rendering capabilities.
♻ ☆ Byzantine-resilient Federated Learning With Adaptivity to Data Heterogeneity
This paper deals with federated learning (FL) in the presence of malicious Byzantine attacks and data heterogeneity. A novel Robust Average Gradient Algorithm (RAGA) is proposed, which leverages the geometric median for aggregation and can freely select the round number for local updating. Different from most existing resilient approaches, which perform convergence analysis based on strongly-convex loss function or homogeneously distributed dataset, we conduct convergence analysis for not only strongly-convex but also non-convex loss function over heterogeneous dataset. According to our theoretical analysis, as long as the fraction of dataset from malicious users is less than half, RAGA can achieve convergence at rate $\mathcal{O}({1}/{T^{2/3- \delta}})$ where $T$ is the iteration number and $\delta \in (0, 2/3)$ for non-convex loss function, and at linear rate for strongly-convex loss function. Moreover, stationary point or global optimal solution is proved to obtainable as data heterogeneity vanishes. Experimental results corroborate the robustness of RAGA to Byzantine attacks and verifies the advantage of RAGA over baselines on convergence performance under various intensity of Byzantine attacks, for heterogeneous dataset.
♻ ☆ Hyacinth6B: A large language model for Traditional Chinese
This research's primary motivation of this study is to address the high hardware and computational demands typically associated with LLMs.Therefore,our goal is to find a balance between model lightness and performance,striving to maximize performance while using a comparatively lightweight model. Hyacinth6B was developed with this objective in mind,aiming to fully leverage the core capabilities of LLMs without incurring substantial resource costs, effectively pushing the boundaries of smaller model's performance. The training approach involves parameter efficient finetuning using the LoRA method.
comment: 14pages
♻ ☆ Large Language Model for Multi-objective Evolutionary Optimization
Multiobjective evolutionary algorithms (MOEAs) are major methods for solving multiobjective optimization problems (MOPs). Many MOEAs have been proposed in the past decades, of which the search operators need a carefully handcrafted design with domain knowledge. Recently, some attempts have been made to replace the manually designed operators in MOEAs with learning-based operators (e.g., neural network models). However, much effort is still required for designing and training such models, and the learned operators might not generalize well on new problems. To tackle the above challenges, this work investigates a novel approach that leverages the powerful large language model (LLM) to design MOEA operators. With proper prompt engineering, we successfully let a general LLM serve as a black-box search operator for decomposition-based MOEA (MOEA/D) in a zero-shot manner. In addition, by learning from the LLM behavior, we further design an explicit white-box operator with randomness and propose a new version of decomposition-based MOEA, termed MOEA/D-LO. Experimental studies on different test benchmarks show that our proposed method can achieve competitive performance with widely used MOEAs. It is also promising to see the operator only learned from a few instances can have robust generalization performance on unseen problems with quite different patterns and settings. The results reveal the potential benefits of using pre-trained LLMs in the design of MOEAs.To foster reproducibility and accessibility, the source code is https://github.com/FeiLiu36/LLM4MOEA.
♻ ☆ DeepMachining: Online Prediction of Machining Errors of Lathe Machines
We describe DeepMachining, a deep learning-based AI system for online prediction of machining errors of lathe machine operations. We have built and evaluated DeepMachining based on manufacturing data from factories. Specifically, we first pretrain a deep learning model for a given lathe machine's operations to learn the salient features of machining states. Then, we fine-tune the pretrained model to adapt to specific machining tasks. We demonstrate that DeepMachining achieves high prediction accuracy for multiple tasks that involve different workpieces and cutting tools. To the best of our knowledge, this work is one of the first factory experiments using pre-trained deep-learning models to predict machining errors of lathe machines.
♻ ☆ Advancing Topic Segmentation and Outline Generation in Chinese Texts: The Paragraph-level Topic Representation, Corpus, and Benchmark LREC
Topic segmentation and outline generation strive to divide a document into coherent topic sections and generate corresponding subheadings, unveiling the discourse topic structure of a document. Compared with sentence-level topic structure, the paragraph-level topic structure can quickly grasp and understand the overall context of the document from a higher level, benefitting many downstream tasks such as summarization, discourse parsing, and information retrieval. However, the lack of large-scale, high-quality Chinese paragraph-level topic structure corpora restrained relative research and applications. To fill this gap, we build the Chinese paragraph-level topic representation, corpus, and benchmark in this paper. Firstly, we propose a hierarchical paragraph-level topic structure representation with three layers to guide the corpus construction. Then, we employ a two-stage man-machine collaborative annotation method to construct the largest Chinese Paragraph-level Topic Structure corpus (CPTS), achieving high quality. We also build several strong baselines, including ChatGPT, to validate the computability of CPTS on two fundamental tasks (topic segmentation and outline generation) and preliminarily verified its usefulness for the downstream task (discourse parsing).
comment: Accepted by LREC-COLING 2024
♻ ☆ ObjectCompose: Evaluating Resilience of Vision-Based Models on Object-to-Background Compositional Changes
Given the large-scale multi-modal training of recent vision-based models and their generalization capabilities, understanding the extent of their robustness is critical for their real-world deployment. In this work, we evaluate the resilience of current vision-based models against diverse object-to-background context variations. The majority of robustness evaluation methods have introduced synthetic datasets to induce changes to object characteristics (viewpoints, scale, color) or utilized image transformation techniques (adversarial changes, common corruptions) on real images to simulate shifts in distributions. Recent works have explored leveraging large language models and diffusion models to generate changes in the background. However, these methods either lack in offering control over the changes to be made or distort the object semantics, making them unsuitable for the task. Our method, on the other hand, can induce diverse object-to-background changes while preserving the original semantics and appearance of the object. To achieve this goal, we harness the generative capabilities of text-to-image, image-to-text, and image-to-segment models to automatically generate a broad spectrum of object-to-background changes. We induce both natural and adversarial background changes by either modifying the textual prompts or optimizing the latents and textual embedding of text-to-image models. We produce various versions of standard vision datasets (ImageNet, COCO), incorporating either diverse and realistic backgrounds into the images or introducing color, texture, and adversarial changes in the background. We conduct extensive experiment to analyze the robustness of vision-based models against object-to-background context variations across diverse tasks. Code https://github.com/Muhammad-Huzaifaa/ObjectCompose.git
♻ ☆ Motion Generation from Fine-grained Textual Descriptions
The task of text2motion is to generate human motion sequences from given textual descriptions, where the model explores diverse mappings from natural language instructions to human body movements. While most existing works are confined to coarse-grained motion descriptions, e.g., "A man squats.", fine-grained descriptions specifying movements of relevant body parts are barely explored. Models trained with coarse-grained texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure to generate motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset specializing in fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with step-by-step instructions with pseudo-code compulsory checks. Accordingly, we design a new text2motion model, FineMotionDiffuse, making full use of fine-grained textual information. Our quantitative evaluation shows that FineMotionDiffuse trained on FineHumanML3D improves FID by a large margin of 0.38, compared with competitive baselines. According to the qualitative evaluation and case study, our model outperforms MotionDiffuse in generating spatially or chronologically composite motions, by learning the implicit mappings from fine-grained descriptions to the corresponding basic motions. We release our data at https://github.com/KunhangL/finemotiondiffuse.
♻ ☆ Tandem Transformers for Inference Efficient LLMs
The autoregressive nature of conventional large language models (LLMs) inherently limits inference speed, as tokens are generated sequentially. While speculative and parallel decoding techniques attempt to mitigate this, they face limitations: either relying on less accurate smaller models for generation or failing to fully leverage the base LLM's representations. We introduce a novel architecture, Tandem transformers, to address these issues. This architecture uniquely combines (1) a small autoregressive model and (2) a large model operating in block mode (processing multiple tokens simultaneously). The small model's predictive accuracy is substantially enhanced by granting it attention to the large model's richer representations. On the PaLM2 pretraining dataset, a tandem of PaLM2-Bison and PaLM2-Gecko demonstrates a 3.3% improvement in next-token prediction accuracy over a standalone PaLM2-Gecko, offering a 1.16x speedup compared to a PaLM2-Otter model with comparable downstream performance. We further incorporate the tandem model within the speculative decoding (SPEED) framework where the large model validates tokens from the small model. This ensures that the Tandem of PaLM2-Bison and PaLM2-Gecko achieves substantial speedup (around 1.14x faster than using vanilla PaLM2-Gecko in SPEED) while maintaining identical downstream task accuracy.
♻ ☆ FedCSD: A Federated Learning Based Approach for Code-Smell Detection
This paper proposes a Federated Learning Code Smell Detection (FedCSD) approach that allows organizations to collaboratively train federated ML models while preserving their data privacy. These assertions have been supported by three experiments that have significantly leveraged three manually validated datasets aimed at detecting and examining different code smell scenarios. In experiment 1, which was concerned with a centralized training experiment, dataset two achieved the lowest accuracy (92.30%) with fewer smells, while datasets one and three achieved the highest accuracy with a slight difference (98.90% and 99.5%, respectively). This was followed by experiment 2, which was concerned with cross-evaluation, where each ML model was trained using one dataset, which was then evaluated over the other two datasets. Results from this experiment show a significant drop in the model's accuracy (lowest accuracy: 63.80\%) where fewer smells exist in the training dataset, which has a noticeable reflection (technical debt) on the model's performance. Finally, the last and third experiments evaluate our approach by splitting the dataset into 10 companies. The ML model was trained on the company's site, then all model-updated weights were transferred to the server. Ultimately, an accuracy of 98.34% was achieved by the global model that has been trained using 10 companies for 100 training rounds. The results reveal a slight difference in the global model's accuracy compared to the highest accuracy of the centralized model, which can be ignored in favour of the global model's comprehensive knowledge, lower training cost, preservation of data privacy, and avoidance of the technical debt problem.
comment: 17 pages, 7 figures, Journal paper
♻ ☆ Deciphering the Impact of Pretraining Data on Large Language Models through Machine Unlearning
Through pretraining on a corpus with various sources, Large Language Models (LLMs) have gained impressive performance. However, the impact of each component of the pretraining corpus remains opaque. As a result, the organization of the pretraining corpus is still empirical and may deviate from the optimal. To address this issue, we systematically analyze the impact of 48 datasets from 5 major categories of pretraining data of LLMs and measure their impacts on LLMs using benchmarks about nine major categories of model capabilities. Our analyses provide empirical results about the contribution of multiple corpora on the performances of LLMs, along with their joint impact patterns, including complementary, orthogonal, and correlational relationships. We also identify a set of ``high-impact data'' such as Books that is significantly related to a set of model capabilities. These findings provide insights into the organization of data to support more efficient pretraining of LLMs.
♻ ☆ High-throughput Biomedical Relation Extraction for Semi-Structured Web Articles Empowered by Large Language Models
Objective: To develop a high-throughput biomedical relation extraction system that takes advantage of the large language models'(LLMs) reading comprehension ability and biomedical world knowledge in a scalable and evidential manner. Methods: We formulate the relation extraction task as binary classifications for large language models. Specifically, LLMs make the decision based on the external corpus and its world knowledge, giving the reason for the judgment for factual verification. This method is tailored for semi-structured web articles, wherein we designate the main title as the tail entity and explicitly incorporate it into the context, and the potential head entities are matched based on a biomedical thesaurus. Moreover, lengthy contents are sliced into text chunks, embedded, and retrieved with additional embedding models. Results: Using an open-source LLM, we extracted 248659 relation triplets of three distinct relation types from three reputable biomedical websites. To assess the efficacy of the basic pipeline employed for biomedical relation extraction, we curated a benchmark dataset annotated by a medical expert. Evaluation results indicate that the pipeline exhibits performance comparable to that of GPT-4. Case studies further illuminate challenges faced by contemporary LLMs in the context of biomedical relation extraction for semi-structured web articles. Conclusion: The proposed method has demonstrated its effectiveness in leveraging the strengths of LLMs for high-throughput biomedical relation extraction. Its adaptability is evident, as it can be seamlessly extended to diverse semi-structured biomedical websites, facilitating the extraction of various types of biomedical relations with ease.
♻ ☆ DreamComposer: Controllable 3D Object Generation via Multi-View Conditions
Utilizing pre-trained 2D large-scale generative models, recent works are capable of generating high-quality novel views from a single in-the-wild image. However, due to the lack of information from multiple views, these works encounter difficulties in generating controllable novel views. In this paper, we present DreamComposer, a flexible and scalable framework that can enhance existing view-aware diffusion models by injecting multi-view conditions. Specifically, DreamComposer first uses a view-aware 3D lifting module to obtain 3D representations of an object from multiple views. Then, it renders the latent features of the target view from 3D representations with the multi-view feature fusion module. Finally the target view features extracted from multi-view inputs are injected into a pre-trained diffusion model. Experiments show that DreamComposer is compatible with state-of-the-art diffusion models for zero-shot novel view synthesis, further enhancing them to generate high-fidelity novel view images with multi-view conditions, ready for controllable 3D object reconstruction and various other applications.
comment: Project Page: https://yhyang-myron.github.io/DreamComposer/
♻ ☆ Attention-based Estimation and Prediction of Human Intent to augment Haptic Glove aided Control of Robotic Hand
The letter focuses on Haptic Glove (HG) based control of a Robotic Hand (RH) executing in-hand manipulation of certain objects of interest. The high dimensional motion signals in HG and RH possess intrinsic variability of kinematics resulting in difficulty to establish a direct mapping of the motion signals from HG onto the RH. An estimation mechanism is proposed to quantify the motion signal acquired from the human controller in relation to the intended goal pose of the object being held by the robotic hand. A control algorithm is presented to transform the synthesized intent at the RH and allow relocation of the object to the expected goal pose. The lag in synthesis of the intent in the presence of communication delay leads to a requirement of predicting the estimated intent. We leverage an attention-based convolutional neural network encoder to predict the trajectory of intent for a certain lookahead to compensate for the delays. The proposed methodology is evaluated across objects of different shapes, mass, and materials. We present a comparative performance of the estimation and prediction mechanisms on 5G-driven real-world robotic setup against benchmark methodologies. The test-MSE in prediction of human intent is reported to yield ~ 97.3 -98.7% improvement of accuracy in comparison to LSTM-based benchmark
♻ ☆ Mastering Text, Code and Math Simultaneously via Fusing Highly Specialized Language Models
Underlying data distributions of natural language, programming code, and mathematical symbols vary vastly, presenting a complex challenge for large language models (LLMs) that strive to achieve high performance across all three domains simultaneously. Achieving a very high level of proficiency for an LLM within a specific domain often requires extensive training with relevant corpora, which is typically accompanied by a sacrifice in performance in other domains. In this paper, we propose to fuse models that are already highly-specialized directly. The proposed fusing framework, UltraFuser, consists of three distinct specialists that are already sufficiently trained on language, coding, and mathematics. A token-level gating mechanism is introduced to blend the specialists' outputs. A two-stage training strategy accompanied by balanced sampling is designed to ensure stability. To effectively train the fused model, we further construct a high-quality supervised instruction tuning dataset, UltraChat 2, which includes text, code, and mathematical content. This dataset comprises approximately 300,000 instructions and covers a wide range of topics in each domain. Experiments show that our model could simultaneously achieve mastery of the three crucial domains.
♻ ☆ Guessing human intentions to avoid dangerous situations in caregiving robots IROS2024
For robots to interact socially, they must interpret human intentions and anticipate their potential outcomes accurately. This is particularly important for social robots designed for human care, which may face potentially dangerous situations for people, such as unseen obstacles in their way, that should be avoided. This paper explores the Artificial Theory of Mind (ATM) approach to inferring and interpreting human intentions. We propose an algorithm that detects risky situations for humans, selecting a robot action that removes the danger in real time. We use the simulation-based approach to ATM and adopt the 'like-me' policy to assign intentions and actions to people. Using this strategy, the robot can detect and act with a high rate of success under time-constrained situations. The algorithm has been implemented as part of an existing robotics cognitive architecture and tested in simulation scenarios. Three experiments have been conducted to test the implementation's robustness, precision and real-time response, including a simulated scenario, a human-in-the-loop hybrid configuration and a real-world scenario.
comment: 8 pages, 6 figures. Submitted to IROS2024. For associated mpeg file see https://youtu.be/87UEB8P97KY
♻ ☆ Navigating the EU AI Act: A Methodological Approach to Compliance for Safety-critical Products
In December 2023, the European Parliament provisionally agreed on the EU AI Act. This unprecedented regulatory framework for AI systems lays out guidelines to ensure the safety, legality, and trustworthiness of AI products. This paper presents a methodology for interpreting the EU AI Act requirements for high-risk AI systems by leveraging product quality models. We first propose an extended product quality model for AI systems, incorporating attributes relevant to the Act not covered by current quality models. We map the Act requirements to relevant quality attributes with the goal of refining them into measurable characteristics. We then propose a contract-based approach to derive technical requirements at the stakeholder level. This facilitates the development and assessment of AI systems that not only adhere to established quality standards, but also comply with the regulatory requirements outlined in the Act for high-risk (including safety-critical) AI systems. We demonstrate the applicability of this methodology on an exemplary automotive supply chain use case, where several stakeholders interact to achieve EU AI Act compliance.
comment: To be published in: 2024 IEEE Conference on Artificial Intelligence (CAI 2024)
♻ ☆ An Implicit GNN Solver for Poisson-like problems
This paper presents $\Psi$-GNN, a novel Graph Neural Network (GNN) approach for solving the ubiquitous Poisson PDE problems with mixed boundary conditions. By leveraging the Implicit Layer Theory, $\Psi$-GNN models an "infinitely" deep network, thus avoiding the empirical tuning of the number of required Message Passing layers to attain the solution. Its original architecture explicitly takes into account the boundary conditions, a critical prerequisite for physical applications, and is able to adapt to any initially provided solution. $\Psi$-GNN is trained using a "physics-informed" loss, and the training process is stable by design, and insensitive to its initialization. Furthermore, the consistency of the approach is theoretically proven, and its flexibility and generalization efficiency are experimentally demonstrated: the same learned model can accurately handle unstructured meshes of various sizes, as well as different boundary conditions. To the best of our knowledge, $\Psi$-GNN is the first physics-informed GNN-based method that can handle various unstructured domains, boundary conditions and initial solutions while also providing convergence guarantees.
♻ ☆ Detecting Sexual Content at the Sentence Level in First Millennium Latin Texts
In this study, we propose to evaluate the use of deep learning methods for semantic classification at the sentence level to accelerate the process of corpus building in the field of humanities and linguistics, a traditional and time-consuming task. We introduce a novel corpus comprising around 2500 sentences spanning from 300 BCE to 900 CE including sexual semantics (medical, erotica, etc.). We evaluate various sentence classification approaches and different input embedding layers, and show that all consistently outperform simple token-based searches. We explore the integration of idiolectal and sociolectal metadata embeddings (centuries, author, type of writing), but find that it leads to overfitting. Our results demonstrate the effectiveness of this approach, achieving high precision and true positive rates (TPR) of respectively 70.60% and 86.33% using HAN. We evaluate the impact of the dataset size on the model performances (420 instead of 2013), and show that, while our models perform worse, they still offer a high enough precision and TPR, even without MLM, respectively 69% and 51%. Given the result, we provide an analysis of the attention mechanism as a supporting added value for humanists in order to produce more data.
♻ ☆ Learning User Embeddings from Human Gaze for Personalised Saliency Prediction
Reusable embeddings of user behaviour have shown significant performance improvements for the personalised saliency prediction task. However, prior works require explicit user characteristics and preferences as input, which are often difficult to obtain. We present a novel method to extract user embeddings from pairs of natural images and corresponding saliency maps generated from a small amount of user-specific eye tracking data. At the core of our method is a Siamese convolutional neural encoder that learns the user embeddings by contrasting the image and personal saliency map pairs of different users. Evaluations on two public saliency datasets show that the generated embeddings have high discriminative power, are effective at refining universal saliency maps to the individual users, and generalise well across users and images. Finally, based on our model's ability to encode individual user characteristics, our work points towards other applications that can benefit from reusable embeddings of gaze behaviour.
♻ ☆ Re2LLM: Reflective Reinforcement Large Language Model for Session-based Recommendation
Large Language Models (LLMs) are emerging as promising approaches to enhance session-based recommendation (SBR), where both prompt-based and fine-tuning-based methods have been widely investigated to align LLMs with SBR. However, the former methods struggle with optimal prompts to elicit the correct reasoning of LLMs due to the lack of task-specific feedback, leading to unsatisfactory recommendations. Although the latter methods attempt to fine-tune LLMs with domain-specific knowledge, they face limitations such as high computational costs and reliance on open-source backbones. To address such issues, we propose a \underline{Re}flective \underline{Re}inforcement \underline{L}arge \underline{L}anguage \underline{M}odel (Re2LLM) for SBR, guiding LLMs to focus on specialized knowledge essential for more accurate recommendations effectively and efficiently. In particular, we first design the Reflective Exploration Module to effectively extract knowledge that is readily understandable and digestible by LLMs. To be specific, we direct LLMs to examine recommendation errors through self-reflection and construct a knowledge base (KB) comprising hints capable of rectifying these errors. To efficiently elicit the correct reasoning of LLMs, we further devise the Reinforcement Utilization Module to train a lightweight retrieval agent. It learns to select hints from the constructed KB based on the task-specific feedback, where the hints can serve as guidance to help correct LLMs reasoning for better recommendations. Extensive experiments on multiple real-world datasets demonstrate that our method consistently outperforms state-of-the-art methods.
comment: 11 pages, 4 figures
♻ ☆ Born With a Silver Spoon? Investigating Socioeconomic Bias in Large Language Models
Socioeconomic bias in society exacerbates disparities, influencing access to opportunities and resources based on individuals' economic and social backgrounds. This pervasive issue perpetuates systemic inequalities, hindering the pursuit of inclusive progress as a society. In this paper, we investigate the presence of socioeconomic bias, if any, in large language models. To this end, we introduce a novel dataset SilverSpoon, consisting of 3000 samples that illustrate hypothetical scenarios that involve underprivileged people performing ethically ambiguous actions due to their circumstances, and ask whether the action is ethically justified. Further, this dataset has a dual-labeling scheme and has been annotated by people belonging to both ends of the socioeconomic spectrum. Using SilverSpoon, we evaluate the degree of socioeconomic bias expressed in large language models and the variation of this degree as a function of model size. We also perform qualitative analysis to analyze the nature of this bias. Our analysis reveals that while humans disagree on which situations require empathy toward the underprivileged, most large language models are unable to empathize with the socioeconomically underprivileged regardless of the situation. To foster further research in this domain, we make SilverSpoon and our evaluation harness publicly available.
♻ ☆ Motion Planning Diffusion: Learning and Planning of Robot Motions with Diffusion Models
Learning priors on trajectory distributions can help accelerate robot motion planning optimization. Given previously successful plans, learning trajectory generative models as priors for a new planning problem is highly desirable. Prior works propose several ways on utilizing this prior to bootstrapping the motion planning problem. Either sampling the prior for initializations or using the prior distribution in a maximum-a-posterior formulation for trajectory optimization. In this work, we propose learning diffusion models as priors. We then can sample directly from the posterior trajectory distribution conditioned on task goals, by leveraging the inverse denoising process of diffusion models. Furthermore, diffusion has been recently shown to effectively encode data multimodality in high-dimensional settings, which is particularly well-suited for large trajectory dataset. To demonstrate our method efficacy, we compare our proposed method - Motion Planning Diffusion - against several baselines in simulated planar robot and 7-dof robot arm manipulator environments. To assess the generalization capabilities of our method, we test it in environments with previously unseen obstacles. Our experiments show that diffusion models are strong priors to encode high-dimensional trajectory distributions of robot motions.
♻ ☆ CLHA: A Simple yet Effective Contrastive Learning Framework for Human Alignment
Reinforcement learning from human feedback (RLHF) is a crucial technique in aligning large language models (LLMs) with human preferences, ensuring these LLMs behave in beneficial and comprehensible ways to users. However, a longstanding challenge in human alignment techniques based on reinforcement learning lies in their inherent complexity and difficulty in training. To address this challenge, we present a simple yet effective Contrastive Learning Framework for Human Alignment (CLHA) to align LLMs with human preferences directly. CLHA employs a novel rescoring strategy to evaluate the noise within the data by considering its inherent quality and dynamically adjusting the training process. Simultaneously, CLHA utilizes pairwise contrastive loss and adaptive supervised fine-tuning loss to adaptively modify the likelihood of generating responses, ensuring enhanced alignment with human preferences. Using advanced methods, CLHA surpasses other algorithms, showcasing superior performance in terms of reward model scores, automatic evaluations, and human assessments on the widely used ``Helpful and Harmless'' dataset.
♻ ☆ Brain Networks and Intelligence: A Graph Neural Network Based Approach to Resting State fMRI Data
Resting-state functional magnetic resonance imaging (rsfMRI) is a powerful tool for investigating the relationship between brain function and cognitive processes as it allows for the functional organization of the brain to be captured without relying on a specific task or stimuli. In this paper, we present a novel modeling architecture called BrainRGIN for predicting intelligence (fluid, crystallized, and total intelligence) using graph neural networks on rsfMRI derived static functional network connectivity matrices. Extending from the existing graph convolution networks, our approach incorporates a clustering-based embedding and graph isomorphism network in the graph convolutional layer to reflect the nature of the brain sub-network organization and efficient network expression, in combination with TopK pooling and attention-based readout functions. We evaluated our proposed architecture on a large dataset, specifically the Adolescent Brain Cognitive Development Dataset, and demonstrated its effectiveness in predicting individual differences in intelligence. Our model achieved lower mean squared errors and higher correlation scores than existing relevant graph architectures and other traditional machine learning models for all of the intelligence prediction tasks. The middle frontal gyrus exhibited a significant contribution to both fluid and crystallized intelligence, suggesting their pivotal role in these cognitive processes. Total composite scores identified a diverse set of brain regions to be relevant which underscores the complex nature of total intelligence.
♻ ☆ NaNa and MiGu: Semantic Data Augmentation Techniques to Enhance Protein Classification in Graph Neural Networks
Protein classification tasks are essential in drug discovery. Real-world protein structures are dynamic, which will determine the properties of proteins. However, the existing machine learning methods, like ProNet (Wang et al., 2022a), only access limited conformational characteristics and protein side-chain features, leading to impractical protein structure and inaccuracy of protein classes in their predictions. In this paper, we propose novel semantic data augmentation methods, Novel Augmentation of New Node Attributes (NaNa), and Molecular Interactions and Geometric Upgrading (MiGu) to incorporate backbone chemical and side-chain biophysical information into protein classification tasks and a co-embedding residual learning framework. Specifically, we leverage molecular biophysical, secondary structure, chemical bonds, and ionic features of proteins to facilitate protein classification tasks. Furthermore, our semantic augmentation methods and the co-embedding residual learning framework can improve the performance of GIN (Xu et al., 2019) on EC and Fold datasets (Bairoch, 2000; Andreeva et al., 2007) by 16.41% and 11.33% respectively. Our code is available at https://github.com/r08b46009/Code_for_MIGU_NANA/tree/main.
♻ ☆ Graph Generation with $K^2$-trees ICLR
Generating graphs from a target distribution is a significant challenge across many domains, including drug discovery and social network analysis. In this work, we introduce a novel graph generation method leveraging $K^2$-tree representation, originally designed for lossless graph compression. The $K^2$-tree representation {encompasses inherent hierarchy while enabling compact graph generation}. In addition, we make contributions by (1) presenting a sequential $K^2$-treerepresentation that incorporates pruning, flattening, and tokenization processes and (2) introducing a Transformer-based architecture designed to generate the sequence by incorporating a specialized tree positional encoding scheme. Finally, we extensively evaluate our algorithm on four general and two molecular graph datasets to confirm its superiority for graph generation.
comment: International Conference on Learning Representations (ICLR) 2024
♻ ☆ A Simple and Scalable Representation for Graph Generation ICLR
Recently, there has been a surge of interest in employing neural networks for graph generation, a fundamental statistical learning problem with critical applications like molecule design and community analysis. However, most approaches encounter significant limitations when generating large-scale graphs. This is due to their requirement to output the full adjacency matrices whose size grows quadratically with the number of nodes. In response to this challenge, we introduce a new, simple, and scalable graph representation named gap encoded edge list (GEEL) that has a small representation size that aligns with the number of edges. In addition, GEEL significantly reduces the vocabulary size by incorporating the gap encoding and bandwidth restriction schemes. GEEL can be autoregressively generated with the incorporation of node positional encoding, and we further extend GEEL to deal with attributed graphs by designing a new grammar. Our findings reveal that the adoption of this compact representation not only enhances scalability but also bolsters performance by simplifying the graph generation process. We conduct a comprehensive evaluation across ten non-attributed and two molecular graph generation tasks, demonstrating the effectiveness of GEEL.
comment: International Conference on Learning Representations (ICLR) 2024
Rumor Detection with a novel graph neural network approach
The wide spread of rumors on social media has caused a negative impact on people's daily life, leading to potential panic, fear, and mental health problems for the public. How to debunk rumors as early as possible remains a challenging problem. Existing studies mainly leverage information propagation structure to detect rumors, while very few works focus on correlation among users that they may coordinate to spread rumors in order to gain large popularity. In this paper, we propose a new detection model, that jointly learns both the representations of user correlation and information propagation to detect rumors on social media. Specifically, we leverage graph neural networks to learn the representations of user correlation from a bipartite graph that describes the correlations between users and source tweets, and the representations of information propagation with a tree structure. Then we combine the learned representations from these two modules to classify the rumors. Since malicious users intend to subvert our model after deployment, we further develop a greedy attack scheme to analyze the cost of three adversarial attacks: graph attack, comment attack, and joint attack. Evaluation results on two public datasets illustrate that the proposed MODEL outperforms the state-of-the-art rumor detection models. We also demonstrate our method performs well for early rumor detection. Moreover, the proposed detection method is more robust to adversarial attacks compared to the best existing method. Importantly, we show that it requires a high cost for attackers to subvert user correlation pattern, demonstrating the importance of considering user correlation for rumor detection.
comment: 10 pages, 5 figures
Image Captioning in news report scenario
Image captioning strives to generate pertinent captions for specified images, situating itself at the crossroads of Computer Vision (CV) and Natural Language Processing (NLP). This endeavor is of paramount importance with far-reaching applications in recommendation systems, news outlets, social media, and beyond. Particularly within the realm of news reporting, captions are expected to encompass detailed information, such as the identities of celebrities captured in the images. However, much of the existing body of work primarily centers around understanding scenes and actions. In this paper, we explore the realm of image captioning specifically tailored for celebrity photographs, illustrating its broad potential for enhancing news industry practices. This exploration aims to augment automated news content generation, thereby facilitating a more nuanced dissemination of information. Our endeavor shows a broader horizon, enriching the narrative in news reporting through a more intuitive image captioning framework.
comment: 10 pages, 4 figures
♻ ☆ X-Portrait: Expressive Portrait Animation with Hierarchical Motion Attention
We propose X-Portrait, an innovative conditional diffusion model tailored for generating expressive and temporally coherent portrait animation. Specifically, given a single portrait as appearance reference, we aim to animate it with motion derived from a driving video, capturing both highly dynamic and subtle facial expressions along with wide-range head movements. As its core, we leverage the generative prior of a pre-trained diffusion model as the rendering backbone, while achieve fine-grained head pose and expression control with novel controlling signals within the framework of ControlNet. In contrast to conventional coarse explicit controls such as facial landmarks, our motion control module is learned to interpret the dynamics directly from the original driving RGB inputs. The motion accuracy is further enhanced with a patch-based local control module that effectively enhance the motion attention to small-scale nuances like eyeball positions. Notably, to mitigate the identity leakage from the driving signals, we train our motion control modules with scaling-augmented cross-identity images, ensuring maximized disentanglement from the appearance reference modules. Experimental results demonstrate the universal effectiveness of X-Portrait across a diverse range of facial portraits and expressive driving sequences, and showcase its proficiency in generating captivating portrait animations with consistently maintained identity characteristics.
♻ ☆ Take Care of Your Prompt Bias! Investigating and Mitigating Prompt Bias in Factual Knowledge Extraction COLING 2024
Recent research shows that pre-trained language models (PLMs) suffer from "prompt bias" in factual knowledge extraction, i.e., prompts tend to introduce biases toward specific labels. Prompt bias presents a significant challenge in assessing the factual knowledge within PLMs. Therefore, this paper aims to improve the reliability of existing benchmarks by thoroughly investigating and mitigating prompt bias. We show that: 1) all prompts in the experiments exhibit non-negligible bias, with gradient-based prompts like AutoPrompt and OptiPrompt displaying significantly higher levels of bias; 2) prompt bias can amplify benchmark accuracy unreasonably by overfitting the test datasets, especially on imbalanced datasets like LAMA. Based on these findings, we propose a representation-based approach to mitigate the prompt bias during inference time. Specifically, we first estimate the biased representation using prompt-only querying, and then remove it from the model's internal representations to generate the debiased representations, which are used to produce the final debiased outputs. Experiments across various prompts, PLMs, and benchmarks show that our approach can not only correct the overfitted performance caused by prompt bias, but also significantly improve the prompt retrieval capability (up to 10% absolute performance gain). These results indicate that our approach effectively alleviates prompt bias in knowledge evaluation, thereby enhancing the reliability of benchmark assessments. Hopefully, our plug-and-play approach can be a golden standard to strengthen PLMs toward reliable knowledge bases. Code and data are released in https://github.com/FelliYang/PromptBias.
comment: Accepted by COLING 2024
♻ ☆ Diffusion Models Generate Images Like Painters: an Analytical Theory of Outline First, Details Later NeurIPS23
How do diffusion generative models convert pure noise into meaningful images? In a variety of pretrained diffusion models (including conditional latent space models like Stable Diffusion), we observe that the reverse diffusion process that underlies image generation has the following properties: (i) individual trajectories tend to be low-dimensional and resemble 2D `rotations'; (ii) high-variance scene features like layout tend to emerge earlier, while low-variance details tend to emerge later; and (iii) early perturbations tend to have a greater impact on image content than later perturbations. To understand these phenomena, we derive and study a closed-form solution to the probability flow ODE for a Gaussian distribution, which shows that the reverse diffusion state rotates towards a gradually-specified target on the image manifold. It also shows that generation involves first committing to an outline, and then to finer and finer details. We find that this solution accurately describes the initial phase of image generation for pretrained models, and can in principle be used to make image generation more efficient by skipping reverse diffusion steps. Finally, we use our solution to characterize the image manifold in Stable Diffusion. Our viewpoint reveals an unexpected similarity between generation by GANs and diffusion and provides a conceptual link between diffusion and image retrieval.
comment: 44 pages, 28 figures. A briefer version was presented at NeurIPS23 Workshop on Diffusion Models [arXiv:2311.10892]
♻ ☆ Can Large Language Models Discern Evidence for Scientific Hypotheses? Case Studies in the Social Sciences
Hypothesis formulation and testing are central to empirical research. A strong hypothesis is a best guess based on existing evidence and informed by a comprehensive view of relevant literature. However, with exponential increase in the number of scientific articles published annually, manual aggregation and synthesis of evidence related to a given hypothesis is a challenge. Our work explores the ability of current large language models (LLMs) to discern evidence in support or refute of specific hypotheses based on the text of scientific abstracts. We share a novel dataset for the task of scientific hypothesis evidencing using community-driven annotations of studies in the social sciences. We compare the performance of LLMs to several state-of-the-art benchmarks and highlight opportunities for future research in this area. The dataset is available at https://github.com/Sai90000/ScientificHypothesisEvidencing.git
♻ ☆ Towards a RAG-based Summarization Agent for the Electron-Ion Collider
The complexity and sheer volume of information encompassing documents, papers, data, and other resources from large-scale experiments demand significant time and effort to navigate, making the task of accessing and utilizing these varied forms of information daunting, particularly for new collaborators and early-career scientists. To tackle this issue, a Retrieval Augmented Generation (RAG)--based Summarization AI for EIC (RAGS4EIC) is under development. This AI-Agent not only condenses information but also effectively references relevant responses, offering substantial advantages for collaborators. Our project involves a two-step approach: first, querying a comprehensive vector database containing all pertinent experiment information; second, utilizing a Large Language Model (LLM) to generate concise summaries enriched with citations based on user queries and retrieved data. We describe the evaluation methods that use RAG assessments (RAGAs) scoring mechanisms to assess the effectiveness of responses. Furthermore, we describe the concept of prompt template-based instruction-tuning which provides flexibility and accuracy in summarization. Importantly, the implementation relies on LangChain, which serves as the foundation of our entire workflow. This integration ensures efficiency and scalability, facilitating smooth deployment and accessibility for various user groups within the Electron Ion Collider (EIC) community. This innovative AI-driven framework not only simplifies the understanding of vast datasets but also encourages collaborative participation, thereby empowering researchers. As a demonstration, a web application has been developed to explain each stage of the RAG Agent development in detail.
comment: updated title to have no latex formatting
♻ ☆ AIOS: LLM Agent Operating System
The integration and deployment of large language model (LLM)-based intelligent agents have been fraught with challenges that compromise their efficiency and efficacy. Among these issues are sub-optimal scheduling and resource allocation of agent requests over the LLM, the difficulties in maintaining context during interactions between agent and LLM, and the complexities inherent in integrating heterogeneous agents with different capabilities and specializations. The rapid increase of agent quantity and complexity further exacerbates these issues, often leading to bottlenecks and sub-optimal utilization of resources. Inspired by these challenges, this paper presents AIOS, an LLM agent operating system, which embeds large language model into operating systems (OS) as the brain of the OS, enabling an operating system "with soul" -- an important step towards AGI. Specifically, AIOS is designed to optimize resource allocation, facilitate context switch across agents, enable concurrent execution of agents, provide tool service for agents, and maintain access control for agents. We present the architecture of such an operating system, outline the core challenges it aims to resolve, and provide the basic design and implementation of the AIOS. Our experiments on concurrent execution of multiple agents demonstrate the reliability and efficiency of our AIOS modules. Through this, we aim to not only improve the performance and efficiency of LLM agents but also to pioneer for better development and deployment of the AIOS ecosystem in the future. The project is open-source at https://github.com/agiresearch/AIOS.
comment: 14 pages, 5 figures, 5 tables; comments and suggestions are appreciated
♻ ☆ Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators
Large Language Models (LLMs) have demonstrated promising capabilities as automatic evaluators in assessing the quality of generated natural language. However, LLMs still exhibit biases in evaluation and often struggle to generate coherent evaluations that align with human assessments. In this work, we first conduct a systematic study of the misalignment between LLM evaluators and human judgement, revealing that existing calibration methods aimed at mitigating biases are insufficient for effectively aligning LLM evaluators. Inspired by the use of preference data in RLHF, we formulate the evaluation as a ranking problem and introduce Pairwise-preference Search (PairS), an uncertainty-guided search method that employs LLMs to conduct pairwise comparisons and efficiently ranks candidate texts. PairS achieves state-of-the-art performance on representative evaluation tasks and demonstrates significant improvements over direct scoring. Furthermore, we provide insights into the role of pairwise preference in quantifying the transitivity of LLMs and demonstrate how PairS benefits from calibration.
♻ ☆ Large Language Models in Biomedical and Health Informatics: A Bibliometric Review
Large Language Models (LLMs) have rapidly become important tools in Biomedical and Health Informatics (BHI), enabling new ways to analyze data, treat patients, and conduct research. This bibliometric review aims to provide a panoramic view of how LLMs have been used in BHI by examining research articles and collaboration networks from 2022 to 2023. It further explores how LLMs can improve Natural Language Processing (NLP) applications in various BHI areas like medical diagnosis, patient engagement, electronic health record management, and personalized medicine. To do this, our bibliometric review identifies key trends, maps out research networks, and highlights major developments in this fast-moving field. Lastly, it discusses the ethical concerns and practical challenges of using LLMs in BHI, such as data privacy and reliable medical recommendations. Looking ahead, we consider how LLMs could further transform biomedical research as well as healthcare delivery and patient outcomes. This bibliometric review serves as a resource for stakeholders in healthcare, including researchers, clinicians, and policymakers, to understand the current state and future potential of LLMs in BHI.
comment: 50 pages, 7 figures, 4 tables
♻ ☆ Enhancing Zero-Shot Chain-of-Thought Reasoning in Large Language Models through Logic COLING 2024
Recent advancements in large language models have showcased their remarkable generalizability across various domains. However, their reasoning abilities still have significant room for improvement, especially when confronted with scenarios requiring multi-step reasoning. Although large language models possess extensive knowledge, their reasoning often fails to effectively utilize this knowledge to establish a coherent thinking paradigm. These models sometimes show hallucinations as their reasoning procedures are unconstrained by logical principles. Aiming at improving the zero-shot chain-of-thought reasoning ability of large language models, we propose LoT (Logical Thoughts), a self-improvement prompting framework that leverages principles rooted in symbolic logic, particularly Reductio ad Absurdum, to systematically verify and rectify the reasoning processes step by step. Experimental evaluations conducted on language tasks in diverse domains, including arithmetic, commonsense, symbolic, causal inference, and social problems, demonstrate the efficacy of enhanced reasoning by logic. The implementation code for LoT can be accessed at: https://github.com/xf-zhao/LoT.
comment: Accepted in COLING 2024. Code see https://github.com/xf-zhao/LoT
♻ ☆ Medical Speech Symptoms Classification via Disentangled Representation
Intent is defined for understanding spoken language in existing works. Both textual features and acoustic features involved in medical speech contain intent, which is important for symptomatic diagnosis. In this paper, we propose a medical speech classification model named DRSC that automatically learns to disentangle intent and content representations from textual-acoustic data for classification. The intent representations of the text domain and the Mel-spectrogram domain are extracted via intent encoders, and then the reconstructed text feature and the Mel-spectrogram feature are obtained through two exchanges. After combining the intent from two domains into a joint representation, the integrated intent representation is fed into a decision layer for classification. Experimental results show that our model obtains an average accuracy rate of 95% in detecting 25 different medical symptoms.
comment: Accepted by the 27th International Conference on Computer Supported Cooperative Work in Design (CSCWD 2024)
♻ ☆ Learning Flexible Body Collision Dynamics with Hierarchical Contact Mesh Transformer ICLR 2024
Recently, many mesh-based graph neural network (GNN) models have been proposed for modeling complex high-dimensional physical systems. Remarkable achievements have been made in significantly reducing the solving time compared to traditional numerical solvers. These methods are typically designed to i) reduce the computational cost in solving physical dynamics and/or ii) propose techniques to enhance the solution accuracy in fluid and rigid body dynamics. However, it remains under-explored whether they are effective in addressing the challenges of flexible body dynamics, where instantaneous collisions occur within a very short timeframe. In this paper, we present Hierarchical Contact Mesh Transformer (HCMT), which uses hierarchical mesh structures and can learn long-range dependencies (occurred by collisions) among spatially distant positions of a body -- two close positions in a higher-level mesh correspond to two distant positions in a lower-level mesh. HCMT enables long-range interactions, and the hierarchical mesh structure quickly propagates collision effects to faraway positions. To this end, it consists of a contact mesh Transformer and a hierarchical mesh Transformer (CMT and HMT, respectively). Lastly, we propose a flexible body dynamics dataset, consisting of trajectories that reflect experimental settings frequently used in the display industry for product designs. We also compare the performance of several baselines using well-known benchmark datasets. Our results show that HCMT provides significant performance improvements over existing methods. Our code is available at https://github.com/yuyudeep/hcmt.
comment: Accepted at ICLR 2024
♻ ☆ Sabiá-2: A New Generation of Portuguese Large Language Models
We introduce Sabi\'a-2, a family of large language models trained on Portuguese texts. The models are evaluated on a diverse range of exams, including entry-level tests for Brazilian universities, professional certification exams, and graduate-level exams for various disciplines such as accounting, economics, engineering, law and medicine. Our results reveal that our best model so far, Sabi\'a-2 Medium, matches or surpasses GPT-4's performance in 23 out of 64 exams and outperforms GPT-3.5 in 58 out of 64 exams. Notably, specialization has a significant impact on a model's performance without the need to increase its size, allowing us to offer Sabi\'a-2 Medium at a price per token that is 10 times cheaper than GPT-4. Finally, we identified that math and coding are key abilities that need improvement.
♻ ☆ HIVE: Harnessing Human Feedback for Instructional Visual Editing CVPR
Incorporating human feedback has been shown to be crucial to align text generated by large language models to human preferences. We hypothesize that state-of-the-art instructional image editing models, where outputs are generated based on an input image and an editing instruction, could similarly benefit from human feedback, as their outputs may not adhere to the correct instructions and preferences of users. In this paper, we present a novel framework to harness human feedback for instructional visual editing (HIVE). Specifically, we collect human feedback on the edited images and learn a reward function to capture the underlying user preferences. We then introduce scalable diffusion model fine-tuning methods that can incorporate human preferences based on the estimated reward. Besides, to mitigate the bias brought by the limitation of data, we contribute a new 1M training dataset, a 3.6K reward dataset for rewards learning, and a 1K evaluation dataset to boost the performance of instructional image editing. We conduct extensive empirical experiments quantitatively and qualitatively, showing that HIVE is favored over previous state-of-the-art instructional image editing approaches by a large margin.
comment: In CVPR, 2024
♻ ☆ Batched Low-Rank Adaptation of Foundation Models
Low-Rank Adaptation (LoRA) has recently gained attention for fine-tuning foundation models by incorporating trainable low-rank matrices, thereby reducing the number of trainable parameters. While LoRA offers numerous advantages, its applicability for real-time serving to a diverse and global user base is constrained by its incapability to handle multiple task-specific adapters efficiently. This imposes a performance bottleneck in scenarios requiring personalized, task-specific adaptations for each incoming request. To mitigate this constraint, we introduce Fast LoRA (FLoRA), a framework in which each input example in a minibatch can be associated with its unique low-rank adaptation weights, allowing for efficient batching of heterogeneous requests. We empirically demonstrate that FLoRA retains the performance merits of LoRA, showcasing competitive results on the MultiPL-E code generation benchmark spanning over 8 languages and a multilingual speech recognition task across 6 languages.
comment: 16 pages, 3 figures
Optimization and Control 51
☆ Optimizing Vaccine Site Locations While Considering Travel Inconvenience and Public Health Outcomes
During the COVID-19 pandemic, there were over three million infections in Los Angeles County (LAC). To facilitate distribution when vaccines first became available, LAC set up six mega-sites for dispensing a large number of vaccines to the public. To understand if another choice of mega-site location would have improved accessibility and health outcomes, and to provide insight into future vaccine allocation problems, we propose a multi-objective mixed integer linear programming model that balances travel convenience, infection reduction, and equitable distribution. We provide a tractable objective formulation that effectively proxies real-world public health goals of reducing infections while considering travel inconvenience and equitable distribution of resources. Compared with the solution empirically used in LAC in 2020, we recommend more dispersed mega-site locations that result in a 28% reduction in travel inconvenience and avert an additional 1,000 infections.
☆ LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning
The machine learning community has witnessed impressive advancements since the first appearance of large language models (LLMs), yet their huge memory consumption has become a major roadblock to large-scale training. Parameter Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA) have been proposed to alleviate this problem, but their performance still fails to match full parameter training in most large-scale fine-tuning settings. Attempting to complement this deficiency, we investigate layerwise properties of LoRA on fine-tuning tasks and observe an uncommon skewness of weight norms across different layers. Utilizing this key observation, a surprisingly simple training strategy is discovered, which outperforms both LoRA and full parameter training in a wide range of settings with memory costs as low as LoRA. We name it Layerwise Importance Sampled AdamW (LISA), a promising alternative for LoRA, which applies the idea of importance sampling to different layers in LLMs and randomly freeze most middle layers during optimization. Experimental results show that with similar or less GPU memory consumption, LISA surpasses LoRA or even full parameter tuning in downstream fine-tuning tasks, where LISA consistently outperforms LoRA by over $11\%$-$37\%$ in terms of MT-Bench scores. On large models, specifically LLaMA-2-70B, LISA achieves on-par or better performance than LoRA on MT-Bench, GSM8K, and PubMedQA, demonstrating its effectiveness across different domains.
☆ The Solution to an Impulse Control Problem Motivated by Optimal Harvesting
We consider a stochastic impulse control problem that is motivated by applications such as the optimal exploitation of a natural resource. In particular, we consider a stochastic system whose uncontrolled state dynamics are modelled by a non-explosive positive linear diffusion. The control that can be applied to this system takes the form of one-sided impulsive action. The objective of the control problem is to maximise a discounted performance criterion that rewards the effect of control action but involves a fixed cost at each time of a control intervention. We derive the complete solution to this problem under general assumptions. It turns out that the solution can take four qualitatively different forms, several of which have not been observed in the literature. In two of the four cases, there exist only $\varepsilon$-optimal control strategies. We also show that the boundary classification of 0 may play a critical role in the solution of the problem. Furthermore, we develop a way for establishing the strong solution to a stochastic impulse control problem's optimally controlled SDE.
☆ Scaling Mixed-Integer Programming for Certification of Neural Network Controllers Using Bounds Tightening
Neural networks offer a computationally efficient approximation of model predictive control, but they lack guarantees on the resulting controlled system's properties. Formal certification of neural networks is crucial for ensuring safety, particularly in safety-critical domains such as autonomous vehicles. One approach to formally certify properties of neural networks is to solve a mixed-integer program based on the network. This approach suffers from scalability issues due to the complexity of solving the resulting mixed-integer programs. Nevertheless, these issues can be (partially) mitigated via bound-tightening techniques prior to forming the mixed-integer program, which results in tighter formulations and faster optimisation. This paper presents bound-tightening techniques in the context of neural network explicit control policies. Bound tightening is particularly important when considering problems spanning multiple time steps of a controlled system, as the bounds must be propagated through the problem depth. Several strategies for bound tightening are evaluated in terms of both computational complexity and tightness of the bounds.
☆ Parallelizable Parametric Nonlinear System Identification via tuning of a Moving Horizon State Estimator
This paper introduces a novel optimization-based approach for parametric nonlinear system identification. Building upon the prediction error method framework, traditionally used for linear system identification, we extend its capabilities to nonlinear systems. The predictions are computed using a moving horizon state estimator with a constant arrival cost. Eventually, both the system parameters and the arrival cost are estimated by minimizing the sum of the squared prediction errors. Since the predictions are induced by the state estimator, the method can be viewed as the tuning of a state estimator, based on its predictive capacities. The present extension of the prediction error method not only enhances performance for nonlinear systems but also enables learning from multiple trajectories with unknown initial states, broadening its applicability in practical scenarios. Additionally, the novel formulation leaves room for the design of efficient and parallelizable optimization algorithms, since each output prediction only depends on a fixed window of past actions and measurements. In the special case of linear time-invariant systems, we show an important property of the proposed method which suggests asymptotic consistency under reasonable assumptions. Numerical examples illustrate the effectiveness and practicality of the approach, and one of the examples also highlights the necessity for the arrival cost.
comment: Submitted to IEEE Control and Decision Conference in March 2024. Contains 7 pages including 5 figures
☆ An Integer Linear Program to create the shifts in a supermarket
The shift design and the personnel scheduling problem is known to be a difficult problem. It is a real-world problem which has lots of applications in the organization of companies. Solutions are usually found by dividing the problem in two steps: first the shifts are created, then the employees are assigned to them by respecting a bunch of constraints. The assignment of different tasks increases the complexity, since we have to consider the skills of the single employee necessary to perform any activity. In this paper we present aa integer linear programming formulation which models together the shift creation and the construction of rosters for employees, with the objective of minimizing the amount of uncovered demand. Finally we provide the results for three real-world instances, confirming that this approach is promising.
☆ Multi Agent Pathfinding for Noise Restricted Hybrid Fuel Unmanned Aerial Vehicles
Multi Agent Path Finding (MAPF) seeks the optimal set of paths for multiple agents from respective start to goal locations such that no paths conflict. We address the MAPF problem for a fleet of hybrid-fuel unmanned aerial vehicles which are subject to location-dependent noise restrictions. We solve this problem by searching a constraint tree for which the subproblem at each node is a set of shortest path problems subject to the noise and fuel constraints and conflict zone avoidance. A labeling algorithm is presented to solve this subproblem, including the conflict zones which are treated as dynamic obstacles. We present the experimental results of the algorithms for various graph sizes and number of agents.
comment: 6 pages, 7 figures
☆ On Structural Non-commutativity in Affine Feedback of SISO Nonlinear Systems
The affine feedback connection of SISO nonlinear systems modeled by Chen--Fliess series is shown to be a group action on the plant which is isomorphic to the semi-direct product of shuffle and additive group of non-commutative formal power series. The additive and multiplicative feedback loops in an affine feedback connection are thus proven to be structurally non-commutative. A flip in the order of these loops results in a net additive feedback loop.
comment: submitted to $26^{th}$ International Symposium on Mathematical Theory of Networks and Systems, 2024
☆ Using quantum computers in control: interval matrix properties
Quantum computing provides a powerful framework for tackling computational problems that are classically intractable. The goal of this paper is to explore the use of quantum computers for solving relevant problems in systems and control theory. In the recent literature, different quantum algorithms have been developed to tackle binary optimization, which plays an important role in various control-theoretic problems. As a prototypical example, we consider the verification of interval matrix properties such as non-singularity and stability on a quantum computer. We present a quantum algorithm solving these problems and we study its performance in simulation. Our results demonstrate that quantum computers provide a promising tool for control whose applicability to further computationally complex problems remains to be explored.
comment: Final version, accepted for publication in Proc. European Control Conference (ECC), 2024
☆ Manifold-Guided Lyapunov Control with Diffusion Models
This paper presents a novel approach to generating stabilizing controllers for a large class of dynamical systems using diffusion models. The core objective is to develop stabilizing control functions by identifying the closest asymptotically stable vector field relative to a predetermined manifold and adjusting the control function based on this finding. To achieve this, we employ a diffusion model trained on pairs consisting of asymptotically stable vector fields and their corresponding Lyapunov functions. Our numerical results demonstrate that this pre-trained model can achieve stabilization over previously unseen systems efficiently and rapidly, showcasing the potential of our approach in fast zero-shot control and generalizability.
comment: 14 pages
☆ Chattering Phenomena in Time-Optimal Control for High-Order Chain-of-Integrators Systems with Full State Constraints
Time-optimal control for high-order chain-of-integrators systems with full state constraints and arbitrary given terminal states remains an open and challenging problem in optimal control theory domain. However, optimal control's behaviors in high-order problems lack of precision characterization, even where the existence of chattering phenomena remain unknown and overlooked. This paper establishes a theoretical framework of chattering phenomena in the problem, focusing on the uniqueness of state constraints inducing chattering, the upper bound of switching times in an unconstrained arc during chattering, and the convergence of states and costates to the chattering limit point. For the first time, this paper proves the existence of chattering phenomena in the problems. The chattering optimal control for 4th order problems with velocity constraints is precisely solved, providing an approach to plan strictly time-optimal snap-limited trajectories, while other cases of order $n\leq4$ are proved to not allow chattering. The conclusions correct the longstanding misconception in the industry regarding the time-optimality of S-shaped trajectories with minimal switching times.
☆ A Globally Convergent Gradient Method with Momentum
In this work, we consider smooth unconstrained optimization problems and we deal with the class of gradient methods with momentum, i.e., descent algorithms where the search direction is defined as a linear combination of the current gradient and the preceding search direction. This family of algorithms includes nonlinear conjugate gradient methods and Polyak's heavy-ball approach, and is thus of high practical and theoretical interest in large-scale nonlinear optimization. We propose a general framework where the scalars of the linear combination defining the search direction are computed simultaneously by minimizing the approximate quadratic model in the 2 dimensional subspace. This strategy allows us to define a class of gradient methods with momentum enjoying global convergence guarantees and an optimal worst-case complexity bound in the nonconvex setting. Differently than all related works in the literature, the convergence conditions are stated in terms of the Hessian matrix of the bi-dimensional quadratic model. To the best of our knowledge, these results are novel to the literature. Moreover, extensive computational experiments show that the gradient methods with momentum here presented outperform classical conjugate gradient methods and are (at least) competitive with the state-of-art method for unconstrained optimization, i.e, L-BFGS method.
☆ An Exact Solution for Allocating Car Parking Spaces on Campus
All over the world, especially in the university environment, planning managers and traffic engineers are constantly faced with the problem of inadequate allocation of car parking spaces to demanded users. Users could either prefer reserved parking spaces to unreserved parking spaces or vice versa. This makes the campus parking manager to be faced with two basic problem which are: the problem of allocating the actual number of available reserved spaces to users without any conflict over the same parking space, and the problem of determining the number of parking permit to be issued for parking lot with unreserved spaces. Hence, an optimal or available solution to the problem is required. This paper investigates a model for allocating car parking spaces, adds a constraint to address the reserved parking policy in a university environment and solves the parking allocation problem using an exact solution method. The result obtained gives the value of the objective function and the optimal allocation of users to each parking lot.
comment: An International Multidiscinary Conference on Research, Development and Practices in Science, Technology, Education, Arts, Management & the Social Science (iSTEAMS). Conference Centre, University of Ibandan, Nigeria. 30 May - 01 June 2013
☆ Enhancing Privacy in Federated Learning through Local Training
In this paper we propose the federated private local training algorithm (Fed-PLT) for federated learning, to overcome the challenges of (i) expensive communications and (ii) privacy preservation. We address (i) by allowing for both partial participation and local training, which significantly reduce the number of communication rounds between the central coordinator and computing agents. The algorithm matches the state of the art in the sense that the use of local training demonstrably does not impact accuracy. Additionally, agents have the flexibility to choose from various local training solvers, such as (stochastic) gradient descent and accelerated gradient descent. Further, we investigate how employing local training can enhance privacy, addressing point (ii). In particular, we derive differential privacy bounds and highlight their dependence on the number of local training epochs. We assess the effectiveness of the proposed algorithm by comparing it to alternative techniques, considering both theoretical analysis and numerical results from a classification task.
☆ Robust Stability for Multiagent Systems with Spatio-Temporally Correlated Packet Loss
A problem with considering correlations in the analysis of multiagent system with stochastic packet loss is that they induce dependencies between agents that are otherwise decoupled, preventing the application of decomposition methods required for efficient evaluation. To circumvent that issue, this paper is proposing an approach based on analysing sets of networks with independent communication links, only considering the correlations in an implicit fashion. Combining ideas from the robust stabilization of Markov jump linear systems with recently proposed techniques for analysing packet loss in multiagent systems, we obtain a linear matrix inequality based stability condition which is independent of the number of agents. The main result is that the set of stabilized probability distributions has non-empty interior such that small correlations cannot lead to instability, even though only distributions of independent links were analysed. Moreover, two examples are provided to demonstrate the applicability of the results to practically relevant scenarios.
comment: 7 pages, 2 figures, 1 table
☆ Nonsmooth convex-concave saddle point problems with cardinality penalties
In this paper, we focus on a class of convexly constrained nonsmooth convex-concave saddle point problems with cardinality penalties. Although such nonsmooth nonconvex-nonconcave and discontinuous min-max problems may not have a saddle point, we show that they have a local saddle point and a global minimax point, and some local saddle points have the lower bound properties. We define a class of strong local saddle points based on the lower bound properties for stability of variable selection. Moreover, we give a framework to construct continuous relaxations of the discontinuous min-max problems based on the convolution, such that they have the same saddle points with the original problem. We also establish the relations between the continuous relaxation problems and the original problems regarding local saddle points, global minimax points, local minimax points and stationary points. Finally, we illustrate our results with distributionally robust sparse convex regression, sparse robust bond portfolio construction and sparse convex-concave logistic regression saddle point problems.
☆ Minimum-Delay Opportunity Charging Scheduling for Electric Buses
Transit agencies that operate battery-electric buses must carefully manage fast-charging infrastructure to extend daily bus range without degrading on-time performance. To support this need, we propose a mixed-integer linear programming model to schedule opportunity charging that minimizes the amount of departure delay in all trips served by electric buses. Our novel approach directly tracks queuing at chargers in order to set and propagate departure delays. Allowing but minimizing delays makes it possible to optimize performance when delays due to traffic conditions and charging needs are inevitable, in contrast with existing methods that require charging to occur during scheduled layover time. To solve the model, we develop two algorithms based on decomposition. The first is an exact solution method based on Combinatorial Benders (CB) decomposition, which avoids directly enumerating the model's logic-based "big M" constraints and their inevitable computational challenges. The second, inspired by the CB approach but more efficient, is a polynomial-time heuristic based on linear programming that we call 3S. Computational experiments on both a simple notional transit network and the real bus system of King County, Washington, USA demonstrate the performance of both methods. The 3S method appears particularly promising for creating good charging schedules quickly at real-world scale.
☆ Quantum Optimization for the Future Energy Grid: Summary and Quantum Utility Prospects
In this project summary paper, we summarize the key results and use-cases explored in the German Federal Ministry of Education and Research (BMBF) funded project "Q-GRID" which aims to assess potential quantum utility optimization applications in the electrical grid. The project focuses on two layers of optimization problems relevant to decentralized energy generation and transmission as well as novel energy transportation/exchange methods such as Peer-2-Peer energy trading and microgrid formation. For select energy grid optimization problems, we demonstrate exponential classical optimizer runtime scaling even for small problem instances, and present initial findings that variational quantum algorithms such as QAOA and hybrid quantum annealing solvers may provide more favourable runtime scaling to obtain similar solution quality. These initial results suggest that quantum computing may be a key enabling technology in the future energy transition insofar that they may be able to solve business problems which are already challenging at small problem instance sizes.
comment: 12 pages. arXiv admin note: text overlap with arXiv:2309.05502
☆ Investigations on Physics-Informed Neural Networks for Aerodynamics
Physics-Informed Neural Networks (PINNs) have recently emerged as a novel approach to simulate complex physical systems on the basis of both data observations and physical models. In this work, we investigate the use of PINNs for various applications in aerodynamics and we explain how to leverage their specific formulation to perform some tasks effectively. In particular, we demonstrate the ability of PINNs to construct parametric surrogate models, to achieve multiphysic couplings and to infer turbulence characteristics via data assimilation. The robustness and accuracy of the PINNs approach are analysed, then current issues and challenges are discussed.
☆ An inexact proximal MM method for a class of nonconvex composite image reconstruction models
This paper concerns a class of composite image reconstruction models for impluse noise removal, which is rather general and covers existing convex and nonconvex models proposed for reconstructing images with impluse noise. For this nonconvex and nonsmooth optimization problem, we propose a proximal majorization-minimization (MM) algorithm with an implementable inexactness criterion by seeking in each step an inexact minimizer of a strongly convex majorization of the objective function, and establish the convergence of the iterate sequence under the KL assumption on the constructed potential function. This inexact proximal MM method is applied to handle gray image deblurring and color image inpainting problems, for which the associated potential function satisfy the required KL assumption. Numerical comparisons with two state-of-art solvers for image deblurring and inpainting tasks validate the efficiency of the proposed algorithm and models.
☆ A Moreau Envelope Approach for LQR Meta-Policy Estimation
We study the problem of policy estimation for the Linear Quadratic Regulator (LQR) in discrete-time linear time-invariant uncertain dynamical systems. We propose a Moreau Envelope-based surrogate LQR cost, built from a finite set of realizations of the uncertain system, to define a meta-policy efficiently adjustable to new realizations. Moreover, we design an algorithm to find an approximate first-order stationary point of the meta-LQR cost function. Numerical results show that the proposed approach outperforms naive averaging of controllers on new realizations of the linear system. We also provide empirical evidence that our method has better sample complexity than Model-Agnostic Meta-Learning (MAML) approaches.
comment: 8 pages
☆ An inexact infeasible arc-search interior-point method for linear programming problems
Inexact interior-point methods (IPMs) are a type of interior-point methods that inexactly solve the linear equation system for obtaining the search direction. On the other hand,arc-search IPMs approximate the central path with an ellipsoidal arc obtained by solving two linear equation systems in each iteration, while conventional line-search IPMs solve one linear system, therefore, the improvement due to the inexact solutions of the linear equation systems can be more beneficial in arc-search IPMs than conventional IPMs. In this paper, we propose an inexact infeasible arc-search interior-point method.We establish that the proposed method is a polynomial-time algorithm through its convergence analysis. The numerical experiments with the conjugate gradient method show that the proposed method can reduce the number of iterations compared to an existing method for benchmark problems; the numbers of iterations are reduced to two-thirds for more than 70% of the problems.
comment: 25 pages, 3 figures
☆ Code Generation for Conic Model-Predictive Control on Microcontrollers with TinyMPC
Conic constraints appear in many important control applications like legged locomotion, robotic manipulation, and autonomous rocket landing. However, current solvers for conic optimization problems have relatively heavy computational demands in terms of both floating-point operations and memory footprint, making them impractical for use on small embedded devices. We extend TinyMPC, an open-source, high-speed solver targeting low-power embedded control applications, to handle second-order cone constraints. We also present code-generation software to enable deployment of TinyMPC on a variety of microcontrollers. We benchmark our generated code against state-of-the-art embedded QP and SOCP solvers, demonstrating a two-order-of-magnitude speed increase over ECOS while consuming less memory. Finally, we demonstrate TinyMPC's efficacy on the Crazyflie, a lightweight, resource-constrained quadrotor with fast dynamics. TinyMPC and its code-generation tools are publicly available at https://tinympc.org.
comment: Submitted to CDC, 2024. First two authors contributed equally
☆ Convergence of Iterative Quadratic Programming for Robust Fixed-Endpoint Transfer of Bilinear Systems
We present a computational method for open-loop minimum-norm control synthesis for fixed-endpoint transfer of bilinear ensemble systems that are indexed by two continuously varying parameters. We suppose that one ensemble parameter scales the homogeneous, linear part of the dynamics, and the second parameter scales the effect of the applied control inputs on the inhomogeneous, bilinear dynamics. This class of dynamical systems is motivated by robust quantum control pulse synthesis, where the ensemble parameters correspond to uncertainty in the free Hamiltonian and inhomogeneity in the control Hamiltonian, respectively. Our computational method is based on polynomial approximation of the ensemble state in parameter space and discretization of the evolution equations in the time domain using a product of matrix exponentials corresponding to zero-order hold controls over the time intervals. The dynamics are successively linearized about control and trajectory iterates to formulate a sequence of quadratic programs for computing perturbations to the control that successively improve the objective until the iteration converges. We use a two-stage computation to first ensure transfer to the desired terminal state, and then minimize the norm of the control function. The method is demonstrated for the canonical uniform transfer problem for the Bloch system that appears in nuclear magnetic resonance, as well as the matter-wave splitting problem for the Raman-Nath system that appears in ultra-cold atom interferometry.
☆ Generalized Maximum Entropy Differential Dynamic Programming
We present a sampling-based trajectory optimization method derived from the maximum entropy formulation of Differential Dynamic Programming with Tsallis entropy. This method can be seen as a generalization of the legacy work with Shannon entropy, which leads to a Gaussian optimal control policy for exploration during optimization. With the Tsallis entropy, the optimal control policy takes the form of $q$-Gaussian, which further encourages exploration with its heavy-tailed shape. Moreover, in our formulation, the exploration variance, which was scaled by a fixed constant inverse temperature in the original formulation with Shannon entropy, is automatically scaled based on the value function of the trajectory. Due to this property, our algorithms can promote exploration when necessary, that is, the cost of the trajectory is high, rather than using the same scaling factor. The simulation results demonstrate the properties of the proposed algorithm described above.
comment: 7 pages, 5 figures, This paper is for CDC 2024
☆ Stochastic Finite Volume Method for Uncertainty Management in Gas Pipeline Network Flows
Natural gas consumption by users of pipeline networks is subject to increasing uncertainty that originates from the intermittent nature of electric power loads serviced by gas-fired generators. To enable computationally efficient optimization of gas network flows subject to uncertainty, we develop a finite volume representation of stochastic solutions of hyperbolic partial differential equation (PDE) systems on graph-connected domains with nodal coupling and boundary conditions. The representation is used to express the physical constraints in stochastic optimization problems for gas flow allocation subject to uncertain parameters. The method is based on the stochastic finite volume approach that was recently developed for uncertainty quantification in transient flows represented by hyperbolic PDEs on graphs. In this study, we develop optimization formulations for steady-state gas flow over actuated transport networks subject to probabilistic constraints. In addition to the distributions for the physical solutions, we examine the dual variables that are produced by way of the optimization, and interpret them as price distributions that quantify the financial volatility that arises through demand uncertainty modeled in an optimization-driven gas market mechanism. We demonstrate the computation and distributional analysis using a single-pipe example and a small test network.
☆ The Flying Sidekick Traveling Salesman Problem with Multiple Drops: A Simple and Effective Heuristic Approach
We study the Flying Sidekick Traveling Salesman Problem with Multiple Drops (FSTSP-MD), a multi-modal last-mile delivery model where a single truck and a single drone cooperatively deliver customer packages. In the FSTSP-MD, the drone can be launched from the truck to deliver multiple packages before it returns to the truck for a new delivery operation. The FSTSP-MD aims to find the synchronized truck and drone delivery routes that minimize the completion time of the delivery process. We develop a simple and effective heuristic approach based on an order-first, split-second scheme. This heuristic combines standard local search and diversification techniques with a novel shortest-path problem that finds FSTSP-MD solutions in polynomial time. We show that our heuristic consistently outperforms state-of-the-art heuristics developed for the FSTSP-MD and the FSTSP (i.e., the single-drop case) through extensive numerical experiments. We also show that the FSTSP-MD substantially reduces completion times compared to a traditional truck-only delivery system. Several managerial insights are described regarding the effects of drone capacity, drone speed, drone flight endurance, and customer distribution.
☆ Deep polytopic autoencoders for low-dimensional linear parameter-varying approximations and nonlinear feedback design
Polytopic autoencoders provide low-dimensional parametrizations of states in a polytope. For nonlinear PDEs, this is readily applied to low-dimensional linear parameter-varying (LPV) approximations as they have been exploited for efficient nonlinear controller design via series expansions of the solution to the state-dependent Riccati equation. In this work, we develop a polytopic autoencoder for control applications and show how it outperforms standard linear approaches in view of LPV approximations of nonlinear systems and how the particular architecture enables higher order series expansions at little extra computational effort. We illustrate the properties and potentials of this approach to computational nonlinear controller design for large-scale systems with a thorough numerical study.
comment: 9 pages, 6 figures, 2 tables
♻ ☆ Minimal covariance realization and system identification algorithm for a class of stochastic linear switched systems with i.i.d. switching
In this paper, we consider stochastic realization theory of Linear Switched Systems (LSS) with i.i.d. switching. We characterize minimality of stochastic LSSs and show existence and uniqueness (up to isomorphism) of minimal LSSs in innovation form. We present a realization algorithm to compute a minimal LSS in innovation form from output and input covariances. Finally, based on this realization algorithm, by replacing true covariances with empirical ones, we propose a statistically consistent system identification algorithm.
♻ ☆ An optimal control perspective on diffusion-based generative modeling NeurIPS 2022
We establish a connection between stochastic optimal control and generative models based on stochastic differential equations (SDEs), such as recently developed diffusion probabilistic models. In particular, we derive a Hamilton-Jacobi-Bellman equation that governs the evolution of the log-densities of the underlying SDE marginals. This perspective allows to transfer methods from optimal control theory to generative modeling. First, we show that the evidence lower bound is a direct consequence of the well-known verification theorem from control theory. Further, we can formulate diffusion-based generative modeling as a minimization of the Kullback-Leibler divergence between suitable measures in path space. Finally, we develop a novel diffusion-based method for sampling from unnormalized densities -- a problem frequently occurring in statistics and computational sciences. We demonstrate that our time-reversed diffusion sampler (DIS) can outperform other diffusion-based sampling approaches on multiple numerical examples.
comment: Accepted for oral presentation at NeurIPS 2022 Workshop on Score-Based Methods
♻ ☆ A randomized algorithm for nonconvex minimization with inexact evaluations and complexity guarantees
We consider minimization of a smooth nonconvex function with inexact oracle access to gradient and Hessian (without assuming access to the function value) to achieve approximate second-order optimality. A novel feature of our method is that if an approximate direction of negative curvature is chosen as the step, we choose its sense to be positive or negative with equal probability. We allow gradients to be inexact in a relative sense and relax the coupling between inexactness thresholds for the first- and second-order optimality conditions. Our convergence analysis includes both an expectation bound based on martingale analysis and a high-probability bound based on concentration inequalities. We apply our algorithm to empirical risk minimization problems and obtain improved gradient sample complexity over existing works.
♻ ☆ Optimal Analysis of Method with Batching for Monotone Stochastic Finite-Sum Variational Inequalities
Variational inequalities are a universal optimization paradigm that is interesting in itself, but also incorporates classical minimization and saddle point problems. Modern realities encourage to consider stochastic formulations of optimization problems. In this paper, we present an analysis of a method that gives optimal convergence estimates for monotone stochastic finite-sum variational inequalities. In contrast to the previous works, our method supports batching and does not lose the oracle complexity optimality. The effectiveness of the algorithm, especially in the case of small but not single batches is confirmed experimentally.
comment: 22 pages, 1 algorithm, 2 figures, 1 table
♻ ☆ Optimal Data Splitting in Distributed Optimization for Machine Learning
The distributed optimization problem has become increasingly relevant recently. It has a lot of advantages such as processing a large amount of data in less time compared to non-distributed methods. However, most distributed approaches suffer from a significant bottleneck - the cost of communications. Therefore, a large amount of research has recently been directed at solving this problem. One such approach uses local data similarity. In particular, there exists an algorithm provably optimally exploiting the similarity property. But this result, as well as results from other works solve the communication bottleneck by focusing only on the fact that communication is significantly more expensive than local computing and does not take into account the various capacities of network devices and the different relationship between communication time and local computing expenses. We consider this setup and the objective of this study is to achieve an optimal ratio of distributed data between the server and local machines for any costs of communications and local computations. The running times of the network are compared between uniform and optimal distributions. The superior theoretical performance of our solutions is experimentally validated.
comment: 17 pages, 2 figures
♻ ☆ Continuous Non-monotone DR-submodular Maximization with Down-closed Convex Constraint
We investigate the continuous non-monotone DR-submodular maximization problem subject to a down-closed convex solvable constraint. Our first contribution is to construct an example to demonstrate that (first-order) stationary points can have arbitrarily bad approximation ratios, and they are usually on the boundary of the feasible domain. These findings are in contrast with the monotone case where any stationary point yields a $1/2$-approximation (Hassani et al. (2017)). Moreover, this example offers insights on how to design improved algorithms by avoiding bad stationary points, such as the restricted continuous local search algorithm (Chekuri et al. (2014)) and the aided measured continuous greedy (Buchbinder and Feldman (2019)). However, the analyses in the last two algorithms only work for the discrete domain because both need to invoke the inequality that the multilinear extension of any submodular set function is bounded from below by its Lovasz extension. Our second contribution, therefore, is to remove this restriction and show that both algorithms can be extended to the continuous domain while retaining the same approximation ratios, and hence offering improved approximation ratios over those in Bian et al. (2017a). for the same problem. At last, we also include numerical experiments to demonstrate our algorithms on problems arising from machine learning and artificial intelligence.
♻ ☆ Activations and Gradients Compression for Model-Parallel Training
Large neural networks require enormous computational clusters of machines. Model-parallel training, when the model architecture is partitioned sequentially between workers, is a popular approach for training modern models. Information compression can be applied to decrease workers communication time, as it is often a bottleneck in such systems. This work explores how simultaneous compression of activations and gradients in model-parallel distributed training setup affects convergence. We analyze compression methods such as quantization and TopK compression, and also experiment with error compensation techniques. Moreover, we employ TopK with AQ-SGD per-batch error feedback approach. We conduct experiments on image classification and language model fine-tuning tasks. Our findings demonstrate that gradients require milder compression rates than activations. We observe that $K=10\%$ is the lowest TopK compression level, which does not harm model convergence severely. Experiments also show that models trained with TopK perform well only when compression is also applied during inference. We find that error feedback techniques do not improve model-parallel training compared to plain compression, but allow model inference without compression with almost no quality drop. Finally, when applied with the AQ-SGD approach, TopK stronger than with $ K=30\%$ worsens model performance significantly.
comment: 17 pages, 6 figures, 5 tables
♻ ☆ A note on weak compactness of occupation measures for an absorbing Markov decision process
We consider an absorbing Markov decision process with a discrete time parameter with Borel state and action spaces. We study the issue of the set of occupation measures being compact for the weak topology and its relation to the control model being uniformly absorbing.
♻ ☆ A robust optimization approach model for a multi-vaccine multi-echelon supply chain
This research investigates a multi-product, multi-echelon, and multi-period vaccine supply chain network model under uncertainty and quality inspection errors. The objective function seeks optimizing the total cost of the supply chain. Moreover, the proposed model is formulated as a mixed integer linear programming problem under multiple sources of uncertain parameters including demand, inspection errors, vaccine waste generated in healthcare centers, and defective treatment rate of vaccine waste. To provide meaningful solutions that are robust against future fluctuation of parameters, the robust optimization approach is utilized to incorporate the decision maker risk attitude under different type of uncertainty sets. Namely, box, polyhedral and combination of interval polyhedral. The performance of the proposed model is demonstrated through an illustrative example. The results show the effect of different types of uncertainties on the overall objective function. Managerial insights and research implications in terms of vaccine supply chain is advised and future research directions are proposed.
♻ ☆ Multi-Objective Optimization for Sparse Deep Multi-Task Learning
Different conflicting optimization criteria arise naturally in various Deep Learning scenarios. These can address different main tasks (i.e., in the setting of Multi-Task Learning), but also main and secondary tasks such as loss minimization versus sparsity. The usual approach is a simple weighting of the criteria, which formally only works in the convex setting. In this paper, we present a Multi-Objective Optimization algorithm using a modified Weighted Chebyshev scalarization for training Deep Neural Networks (DNNs) with respect to several tasks. By employing this scalarization technique, the algorithm can identify all optimal solutions of the original problem while reducing its complexity to a sequence of single-objective problems. The simplified problems are then solved using an Augmented Lagrangian method, enabling the use of popular optimization techniques such as Adam and Stochastic Gradient Descent, while efficaciously handling constraints. Our work aims to address the (economical and also ecological) sustainability issue of DNN models, with a particular focus on Deep Multi-Task models, which are typically designed with a very large number of weights to perform equally well on multiple tasks. Through experiments conducted on two Machine Learning datasets, we demonstrate the possibility of adaptively sparsifying the model during training without significantly impacting its performance, if we are willing to apply task-specific adaptations to the network weights. Code is available at https://github.com/salomonhotegni/MDMTN
comment: 12 pages, 7 figures
♻ ☆ Frequency Regulation with Storage: On Losses and Profits
Low-carbon societies will need to store vast amounts of electricity to balance intermittent generation from wind and solar energy, for example, through frequency regulation. Here, we derive an analytical solution to the decision-making problem of storage operators who sell frequency regulation power to grid operators and trade electricity on day-ahead markets. Mathematically, we treat future frequency deviation trajectories as functional uncertainties in a receding horizon robust optimization problem. We constrain the expected terminal state-of-charge to be equal to some target to allow storage operators to make good decisions not only for the present but also the future. Thanks to this constraint, the amount of electricity traded on day-ahead markets is an implicit function of the regulation power sold to grid operators. The implicit function quantifies the amount of power that needs to be purchased to cover the expected energy loss that results from providing frequency regulation. We show how the marginal cost associated with the expected energy loss decreases with roundtrip efficiency and increases with frequency deviation dispersion. We find that the profits from frequency regulation over the lifetime of energy-constrained storage devices are roughly inversely proportional to the length of time for which regulation power must be committed.
♻ ☆ Dual Conic Proxies for AC Optimal Power Flow
In recent years, there has been significant interest in the development of machine learning-based optimization proxies for AC Optimal Power Flow (AC-OPF). Although significant progress has been achieved in predicting high-quality primal solutions, no existing learning-based approach can provide valid dual bounds for AC-OPF. This paper addresses this gap by training optimization proxies for a convex relaxation of AC-OPF. Namely, the paper considers a second-order cone (SOC) relaxation of AC-OPF, and proposes \revision{a novel architecture} that embeds a fast, differentiable (dual) feasibility recovery, thus providing valid dual bounds. The paper combines this new architecture with a self-supervised learning scheme, which alleviates the need for costly training data generation. Extensive numerical experiments on medium- and large-scale power grids demonstrate the efficiency and scalability of the proposed methodology.
comment: accepted to PSCC 2024
♻ ☆ On the Post-Lie Structure in SISO Affine Feedback Control Systems
The main objective of this work is to show that the single-input, single-output (SISO) affine feedback group, a transformation group in the context of the affine feedback interconnection of Chen-Fliess series, is a post-group in the sense of Bai, Guo, Sheng and Tang.
♻ ☆ Optimal Design of Volt/VAR Control Rules of Inverters using Deep Learning
Distribution grids are challenged by rapid voltage fluctuations induced by variable power injections from distributed energy resources (DERs). To regulate voltage, the IEEE Standard 1547 recommends each DER inject reactive power according to piecewise-affine Volt/VAR control rules. Although the standard suggests a default shape, the rule can be customized per bus. This task of optimal rule design (ORD) is challenging as Volt/VAR rules introduce nonlinear dynamics, and lurk trade-offs between stability and steady-state voltage profiles. ORD is formulated as a mixed-integer nonlinear program (MINLP), but scales unfavorably with the problem size. Towards a more efficient solution, we reformulate ORD as a deep learning problem. The idea is to design a DNN that emulates Volt/VAR dynamics. The DNN takes grid scenarios as inputs, rule parameters as weights, and outputs equilibrium voltages. Optimal rule parameters can be found by training the DNN so its output approaches unity for various scenarios. The DNN is only used to optimize rules and is never employed in the field. While dealing with ORD, we also review and expand on stability conditions and convergence rates for Volt/VAR dynamics on single- and multi-phase feeders. Tests showcase the merit of DNN-based ORD by benchmarking it against its MINLP counterpart.
comment: Accepted in the IEEE Trans. on Smart Grid
♻ ☆ Harmonic Control Lyapunov Barrier Functions for Constrained Optimal Control with Reach-Avoid Specifications
This paper introduces harmonic control Lyapunov barrier functions (harmonic CLBF) that aid in constrained control problems such as reach-avoid problems. Harmonic CLBFs exploit the maximum principle that harmonic functions satisfy to encode the properties of control Lyapunov barrier functions (CLBFs). As a result, they can be initiated at the start of an experiment rather than trained based on sample trajectories. The control inputs are selected to maximize the inner product of the system dynamics with the steepest descent direction of the harmonic CLBF. Numerical results are presented with four different systems under different reach-avoid environments. Harmonic CLBFs show a significantly low risk of entering unsafe regions and a high probability of entering the goal region.
♻ ☆ Gaussian smoothing gradient descent for minimizing functions (GSmoothGD)
This work analyzes the convergence of a class of smoothing-based gradient descent methods when applied to optimization problems. In particular, Gaussian smoothing is employed to define a nonlocal gradient that reduces high-frequency noise, small variations, and rapid fluctuations in the computation of the descent directions while preserving the structure and features of the loss landscape. The resulting Gaussian smoothing gradient descent (GSmoothGD) approach can facilitate gradient descent in navigating away from and avoiding local minima with increased ease, thereby substantially enhancing its overall performance even when applied to non-convex optimization problems. This work also provides rigorous theoretical error estimates on the rate of convergence of GSmoothGD iterates. These estimates exemplify the impact of underlying function convexity, smoothness, input dimension, and the Gaussian smoothing radius. To combat the curse of dimensionality, we numerically approximate the GSmoothGD nonlocal gradient using Monte Carlo (MC) sampling and provide a theory in which the iterates converge regardless of the function smoothness and dimension. Finally, we present several strategies to update the smoothing parameter aimed at diminishing the impact of local minima, thereby rendering the attainment of global minima more achievable. Computational evidence complements the present theory and shows the effectiveness of the MC-GSmoothGD method compared to other smoothing-based algorithms, momentum-based approaches, and classical gradient-based algorithms from numerical optimization.
comment: 29 pages, 2 figures, 2 tables
♻ ☆ Discretized Distributed Optimization over Dynamic Digraphs
We consider a discrete-time model of continuous-time distributed optimization over dynamic directed-graphs (digraphs) with applications to distributed learning. Our optimization algorithm works over general strongly connected dynamic networks under switching topologies, e.g., in mobile multi-agent systems and volatile networks due to link failures. Compared to many existing lines of work, there is no need for bi-stochastic weight designs on the links. The existing literature mostly needs the link weights to be stochastic using specific weight-design algorithms needed both at the initialization and at all times when the topology of the network changes. This paper eliminates the need for such algorithms and paves the way for distributed optimization over time-varying digraphs. We derive the bound on the gradient-tracking step-size and discrete time-step for convergence and prove dynamic stability using arguments from consensus algorithms, matrix perturbation theory, and Lyapunov theory. This work, particularly, is an improvement over existing stochastic-weight undirected networks in case of link removal or packet drops. This is because the existing literature may need to rerun time-consuming and computationally complex algorithms for stochastic design, while the proposed strategy works as long as the underlying network is weight-symmetric and balanced. The proposed optimization framework finds applications to distributed classification and learning.
♻ ☆ A Semismooth Newton Stochastic Proximal Point Algorithm with Variance Reduction
We develop an implementable stochastic proximal point (SPP) method for a class of weakly convex, composite optimization problems. The proposed stochastic proximal point algorithm incorporates a variance reduction mechanism and the resulting SPP updates are solved using an inexact semismooth Newton framework. We establish detailed convergence results that take the inexactness of the SPP steps into account and that are in accordance with existing convergence guarantees of (proximal) stochastic variance-reduced gradient methods. Numerical experiments show that the proposed algorithm competes favorably with other state-of-the-art methods and achieves higher robustness with respect to the step size selection.
♻ ☆ On the Geometric Convergence of Byzantine-Resilient Distributed Optimization Algorithms
The problem of designing distributed optimization algorithms that are resilient to Byzantine adversaries has received significant attention. For the Byzantine-resilient distributed optimization problem, the goal is to (approximately) minimize the average of the local cost functions held by the regular (non adversarial) agents in the network. In this paper, we provide a general algorithmic framework for Byzantine-resilient distributed optimization which includes some state-of-the-art algorithms as special cases. We analyze the convergence of algorithms within the framework, and derive a geometric rate of convergence of all regular agents to a ball around the optimal solution (whose size we characterize). Furthermore, we show that approximate consensus can be achieved geometrically fast under some minimal conditions. Our analysis provides insights into the relationship among the convergence region, distance between regular agents' values, step-size, and properties of the agents' functions for Byzantine-resilient distributed optimization.
comment: 29 pages, 5 figures
♻ ☆ Omega: Optimistic EMA Gradients ICML 2023
Stochastic min-max optimization has gained interest in the machine learning community with the advancements in GANs and adversarial training. Although game optimization is fairly well understood in the deterministic setting, some issues persist in the stochastic regime. Recent work has shown that stochastic gradient descent-ascent methods such as the optimistic gradient are highly sensitive to noise or can fail to converge. Although alternative strategies exist, they can be prohibitively expensive. We introduce Omega, a method with optimistic-like updates that mitigates the impact of noise by incorporating an EMA of historic gradients in its update rule. We also explore a variation of this algorithm that incorporates momentum. Although we do not provide convergence guarantees, our experiments on stochastic games show that Omega outperforms the optimistic gradient method when applied to linear players.
comment: Oral at the LatinX in AI workshop @ ICML 2023
♻ ☆ Polyamorous Scheduling
Finding schedules for pairwise meetings between the members of a complex social group without creating interpersonal conflict is challenging, especially when different relationships have different needs. We formally define and study the underlying optimisation problem: Polyamorous Scheduling. In Polyamorous Scheduling, we are given an edge-weighted graph and try to find a periodic schedule of matchings in this graph such that the maximal weighted waiting time between consecutive occurrences of the same edge is minimised. We show that the problem is NP-hard and that there is no efficient approximation algorithm with a better ratio than 4/3 unless P = NP. On the positive side, we obtain an $O(\log n)$-approximation algorithm; indeed, a $O(\log \Delta)$-approximation for $\Delta$ the maximum degree, i.e., the largest number of relationships of any individual. We also define a generalisation of density from the Pinwheel Scheduling Problem, "poly density", and ask whether there exists a poly-density threshold similar to the 5/6-density threshold for Pinwheel Scheduling [Kawamura, STOC 2024]. Polyamorous Scheduling is a natural generalisation of Pinwheel Scheduling with respect to its optimisation variant, Bamboo Garden Trimming. Our work contributes the first nontrivial hardness-of-approximation reduction for any periodic scheduling problem, and opens up numerous avenues for further study of Polyamorous Scheduling.
comment: v2: stronger and simplified hardness-of-approximation results, corrected constant in layering approximation algorithm
♻ ☆ Artificial Intelligence for Operations Research: Revolutionizing the Operations Research Process
The rapid advancement of artificial intelligence (AI) techniques has opened up new opportunities to revolutionize various fields, including operations research (OR). This survey paper explores the integration of AI within the OR process (AI4OR) to enhance its effectiveness and efficiency across multiple stages, such as parameter generation, model formulation, and model optimization. By providing a comprehensive overview of the state-of-the-art and examining the potential of AI to transform OR, this paper aims to inspire further research and innovation in the development of AI-enhanced OR methods and tools. The synergy between AI and OR is poised to drive significant advancements and novel solutions in a multitude of domains, ultimately leading to more effective and efficient decision-making.
♻ ☆ Approximate and Weighted Data Reconstruction Attack in Federated Learning
Federated Learning (FL) is a distributed learning paradigm that enables multiple clients to collaborate on building a machine learning model without sharing their private data. Although FL is considered privacy-preserved by design, recent data reconstruction attacks demonstrate that an attacker can recover clients' training data based on the parameters shared in FL. However, most existing methods fail to attack the most widely used horizontal Federated Averaging (FedAvg) scenario, where clients share model parameters after multiple local training steps. To tackle this issue, we propose an interpolation-based approximation method, which makes attacking FedAvg scenarios feasible by generating the intermediate model updates of the clients' local training processes. Then, we design a layer-wise weighted loss function to improve the data quality of reconstruction. We assign different weights to model updates in different layers concerning the neural network structure, with the weights tuned by Bayesian optimization. Finally, experimental results validate the superiority of our proposed approximate and weighted attack (AWA) method over the other state-of-the-art methods, as demonstrated by the substantial improvement in different evaluation metrics for image data reconstructions.
Systems and Control 48
☆ Multi-Agent Clarity-Aware Dynamic Coverage with Gaussian Processes
This paper presents two algorithms for multi-agent dynamic coverage in spatiotemporal environments, where the coverage algorithms are informed by the method of data assimilation. In particular, we show that by considering the information assimilation algorithm, here a Numerical Gaussian Process Kalman Filter, the influence of measurements taken at one position on the uncertainty of the estimate at another location can be computed. We use this relationship to propose new coverage algorithms. Furthermore, we show that the controllers naturally extend to the multi-agent context, allowing for a distributed-control central-information paradigm for multi-agent coverage. Finally, we demonstrate the algorithms through a realistic simulation of a team of UAVs collecting wind data over a region in Austria.
comment: 8 pages, 2 figures, submitted to CDC 2024
☆ Multi-Agent Resilient Consensus under Intermittent Faulty and Malicious Transmissions (Extended Version)
In this work, we consider the consensus problem in which legitimate agents share their values over an undirected communication network in the presence of malicious or faulty agents. Different from the previous works, we characterize the conditions that generalize to several scenarios such as intermittent faulty or malicious transmissions, based on trust observations. As the standard trust aggregation approach based on a constant threshold fails to distinguish intermittent malicious/faulty activity, we propose a new detection algorithm utilizing time-varying thresholds and the random trust values available to legitimate agents. Under these conditions, legitimate agents almost surely determine their trusted neighborhood correctly with geometrically decaying misclassification probabilities. We further prove that the consensus process converges almost surely even in the presence of malicious agents. We also derive the probabilistic bounds on the deviation from the nominal consensus value that would have been achieved with no malicious agents in the system. Numerical results verify the convergence among agents and exemplify the deviation under different scenarios.
comment: Extended Version of CDC '24 submission
☆ Stealthy Deactivation of Safety Filters
Safety filters ensure that only safe control actions are executed. We propose a simple and stealthy false-data injection attack for deactivating such safety filters; in particular, we focus on deactivating safety filters that are based on control-barrier functions. The attack injects false sensor measurements to bias state estimates to the interior of a safety region, which makes the safety filter accept unsafe control actions. To detect such attacks, we also propose a detector that detects biases manufactured by the proposed attack policy, which complements conventional detectors when safety filters are used. The proposed attack policy and detector are illustrated on a double integrator example.
comment: ECC24
☆ Learning the Optimal Power Flow: Environment Design Matters
To solve the optimal power flow (OPF) problem, reinforcement learning (RL) emerges as a promising new approach. However, the RL-OPF literature is strongly divided regarding the exact formulation of the OPF problem as an RL environment. In this work, we collect and implement diverse environment design decisions from the literature regarding training data, observation space, episode definition, and reward function choice. In an experimental analysis, we show the significant impact of these environment design options on RL-OPF training performance. Further, we derive some first recommendations regarding the choice of these design decisions. The created environment framework is fully open-source and can serve as a benchmark for future research in the RL-OPF field.
☆ Steering Feedback in Dynamic Driving Simulators: The Influence of Steering Wheel Vibration and Vehicle Motion Frequency
The validity of the subjective evaluation of steering feedback in driving simulators is crucial for modern vehicle development. Although there are established objective steering characteristics for the assessment of both stationary and dynamic feedback behaviour, factors such as steering wheel vibrations and vehicle body motion, particularly in high-frequency ranges, present challenges in simulator fidelity. This work investigates the influence of steering wheel vibration and vehicle body motion frequency content on the subjective evaluation of steering feedback during closed-loop driving in a dynamic driving simulator. A controlled subject study with 30 participants consisting of a back-to-back comparison of a reference vehicle with an electrical power steering system and three variants of its virtual representation on a dynamic driving simulator was performed. Subjective evaluation focused on the representation of road feedback in comparison to the reference vehicle. The statistical analysis of subjective results show that there is a significant influence of the frequency content of both steering wheel torque and vehicle motion on the subjective evaluation of steering feedback in a dynamic driving simulator. The results suggest an influence of frequency content on the subjective evaluation quality of steering feedback characteristics that are not associated with the dynamic feedback behaviour in the context of established performance indicators.
comment: 12 pages, 7 figures, 9 tables, submitted to the IEEE Transactions on Intelligent Vehicles
☆ Neural Exponential Stabilization of Control-affine Nonlinear Systems
This paper proposes a novel learning-based approach for achieving exponential stabilization of nonlinear control-affine systems. We leverage the Control Contraction Metrics (CCMs) framework to co-synthesize Neural Contraction Metrics (NCMs) and Neural Network (NN) controllers. First, we transform the infinite-dimensional semi-definite program (SDP) for CCM computation into a tractable inequality feasibility problem using element-wise bounds of matrix-valued functions. The terms in the inequality can be efficiently computed by our novel algorithms. Second, we propose a free parametrization of NCMs guaranteeing positive definiteness and the satisfaction of a partial differential equation, regardless of trainable parameters. Third, this parametrization and the inequality condition enable the design of contractivity-enforcing regularizers, which can be incorporated while designing the NN controller for exponential stabilization of the underlying nonlinear systems. Furthermore, when the training loss goes to zero, we provide formal guarantees on verification of the NCM and the exponentional stabilization under the NN controller. Finally, we validate our method through benchmark experiments on set-point stabilization and increasing the region of attraction of a locally pre-stabilized closed-loop system.
comment: This paper is submitted in CDC2024 for a possible publication
☆ A PAC-Bayesian Framework for Optimal Control with Stability Guarantees
Stochastic Nonlinear Optimal Control (SNOC) involves minimizing a cost function that averages out the random uncertainties affecting the dynamics of nonlinear systems. For tractability reasons, this problem is typically addressed by minimizing an empirical cost, which represents the average cost across a finite dataset of sampled disturbances. However, this approach raises the challenge of quantifying the control performance against out-of-sample uncertainties. Particularly, in scenarios where the training dataset is small, SNOC policies are prone to overfitting, resulting in significant discrepancies between the empirical cost and the true cost, i.e., the average SNOC cost incurred during control deployment. Therefore, establishing generalization bounds on the true cost is crucial for ensuring reliability in real-world applications. In this paper, we introduce a novel approach that leverages PAC-Bayes theory to provide rigorous generalization bounds for SNOC. Based on these bounds, we propose a new method for designing optimal controllers, offering a principled way to incorporate prior knowledge into the synthesis process, which aids in improving the control policy and mitigating overfitting. Furthermore, by leveraging recent parametrizations of stabilizing controllers for nonlinear systems, our framework inherently ensures closed-loop stability. The effectiveness of our proposed method in incorporating prior knowledge and combating overfitting is shown by designing neural network controllers for tasks in cooperative robotics.
☆ Neural Distributed Controllers with Port-Hamiltonian Structures
Controlling large-scale cyber-physical systems necessitates optimal distributed policies, relying solely on local real-time data and limited communication with neighboring agents. However, finding optimal controllers remains challenging, even in seemingly simple scenarios. Parameterizing these policies using Neural Networks (NNs) can deliver good performance, but their sensitivity to small input changes can destabilize the closed-loop system. This paper addresses this issue for a network of nonlinear dissipative systems. Specifically, we leverage well-established port-Hamiltonian structures to characterize deep distributed control policies with closed-loop stability guarantees and a finite $\mathcal{L}_2$ gain, regardless of specific NN parameters. This eliminates the need to constrain the parameters during optimization and enables training with standard methods like stochastic gradient descent. A numerical study on the consensus control of Kuramoto oscillators demonstrates the effectiveness of the proposed controllers.
comment: This paper is submitted in CDC2024 for a possible publication
☆ Optical Flow Based Detection and Tracking of Moving Objects for Autonomous Vehicles
Accurate velocity estimation of surrounding moving objects and their trajectories are critical elements of perception systems in Automated/Autonomous Vehicles (AVs) with a direct impact on their safety. These are non-trivial problems due to the diverse types and sizes of such objects and their dynamic and random behaviour. Recent point cloud based solutions often use Iterative Closest Point (ICP) techniques, which are known to have certain limitations. For example, their computational costs are high due to their iterative nature, and their estimation error often deteriorates as the relative velocities of the target objects increase (>2 m/sec). Motivated by such shortcomings, this paper first proposes a novel Detection and Tracking of Moving Objects (DATMO) for AVs based on an optical flow technique, which is proven to be computationally efficient and highly accurate for such problems. \textcolor{black}{This is achieved by representing the driving scenario as a vector field and applying vector calculus theories to ensure spatiotemporal continuity.} We also report the results of a comprehensive performance evaluation of the proposed DATMO technique, carried out in this study using synthetic and real-world data. The results of this study demonstrate the superiority of the proposed technique, compared to the DATMO techniques in the literature, in terms of estimation accuracy and processing time in a wide range of relative velocities of moving objects. Finally, we evaluate and discuss the sensitivity of the estimation error of the proposed DATMO technique to various system and environmental parameters, as well as the relative velocities of the moving objects.
comment: This manuscript has been accepted as a regular paper in Transactions on Intelligent Transportation Systems (DOI: 10.1109/TITS.2024.3382495)
☆ On Structural Non-commutativity in Affine Feedback of SISO Nonlinear Systems
The affine feedback connection of SISO nonlinear systems modeled by Chen--Fliess series is shown to be a group action on the plant which is isomorphic to the semi-direct product of shuffle and additive group of non-commutative formal power series. The additive and multiplicative feedback loops in an affine feedback connection are thus proven to be structurally non-commutative. A flip in the order of these loops results in a net additive feedback loop.
comment: submitted to $26^{th}$ International Symposium on Mathematical Theory of Networks and Systems, 2024
☆ Using quantum computers in control: interval matrix properties
Quantum computing provides a powerful framework for tackling computational problems that are classically intractable. The goal of this paper is to explore the use of quantum computers for solving relevant problems in systems and control theory. In the recent literature, different quantum algorithms have been developed to tackle binary optimization, which plays an important role in various control-theoretic problems. As a prototypical example, we consider the verification of interval matrix properties such as non-singularity and stability on a quantum computer. We present a quantum algorithm solving these problems and we study its performance in simulation. Our results demonstrate that quantum computers provide a promising tool for control whose applicability to further computationally complex problems remains to be explored.
comment: Final version, accepted for publication in Proc. European Control Conference (ECC), 2024
☆ Prioritize Team Actions: Multi-Agent Temporal Logic Task Planning with Ordering Constraints
In this paper, we investigate the problem of linear temporal logic (LTL) path planning for multi-agent systems, introducing the new concept of \emph{ordering constraints}. Specifically, we consider a generic objective function that is defined for the path of each individual agent. The primary objective is to find a global plan for the team of agents, ensuring they collectively meet the specified LTL requirements. Simultaneously, we aim to maintain a pre-determined order in the values of the objective function for each agent, which we refer to as the ordering constraints. This new requirement stems from scenarios like security-aware planning, where relative orders outweigh absolute values in importance. We present an efficient algorithm to solve this problem, supported by proofs of correctness that demonstrate the optimality of our solution. Additionally, we provide a case study in security-aware path planning to illustrate the practicality and effectiveness of our proposed approach.
☆ Chattering Phenomena in Time-Optimal Control for High-Order Chain-of-Integrators Systems with Full State Constraints
Time-optimal control for high-order chain-of-integrators systems with full state constraints and arbitrary given terminal states remains an open and challenging problem in optimal control theory domain. However, optimal control's behaviors in high-order problems lack of precision characterization, even where the existence of chattering phenomena remain unknown and overlooked. This paper establishes a theoretical framework of chattering phenomena in the problem, focusing on the uniqueness of state constraints inducing chattering, the upper bound of switching times in an unconstrained arc during chattering, and the convergence of states and costates to the chattering limit point. For the first time, this paper proves the existence of chattering phenomena in the problems. The chattering optimal control for 4th order problems with velocity constraints is precisely solved, providing an approach to plan strictly time-optimal snap-limited trajectories, while other cases of order $n\leq4$ are proved to not allow chattering. The conclusions correct the longstanding misconception in the industry regarding the time-optimality of S-shaped trajectories with minimal switching times.
☆ Ultrafast Adaptive Primary Frequency Tuning and Secondary Frequency Identification for S/S WPT system
Magnetic resonance wireless power transfer (WPT) technology is increasingly being adopted across diverse applications. However, its effectiveness can be significantly compromised by parameter shifts within the resonance network, owing to its high system quality factor. Such shifts are inherent and challenging to mitigate during the manufacturing process. In response, this article introduces a rapid frequency tuning approach. Leveraging switch-controlled capacitors (SCC) to adjust the resonance network and the primary side's operating frequency, alongside a current zero-crossing detection (ZCD) circuit for voltage-current phase determination, this method circumvents the need for intricate knowledge of WPT system parameters. Moreover, it obviates the necessity for inter-side communication for real-time identification of the secondary side resonance frequency. The swift response of SCC and two-step perturb-and-observe algorithm mitigate output disturbances, thereby expediting the frequency tuning process. Experimental validation on a 200W Series-Series compensated WPT (SS-WPT) system demonstrates that the proposed method achieves frequency recognition accuracy within 0.7kHz in less than 1ms, increasing system efficiency up to 9%.
comment: 11 pages,16 figures,to be published in IEEE Transactions on Industrial Electronics
☆ Aerial Robots Carrying Flexible Cables: Dynamic Shape Optimal Control via Spectral Method Model
In this work, we present a model-based optimal boundary control design for an aerial robotic system composed of a quadrotor carrying a flexible cable. The whole system is modeled by partial differential equations (PDEs) combined with boundary conditions described by ordinary differential equations (ODEs). The proper orthogonal decomposition (POD) method is adopted to project the original infinite-dimensional system on a subspace spanned by orthogonal basis functions. Based on the reduced order model, nonlinear model predictive control (NMPC) is implemented online to realize shape trajectory tracking of the flexible cable in an optimal predictive fashion. The proposed reduced modeling and optimal control paradigms are numerically verified against an accurate high-dimensional FDM-based model in different scenarios and the controller's superior performance is shown compared to an optimally tuned PID controller.
☆ Robust Stability for Multiagent Systems with Spatio-Temporally Correlated Packet Loss
A problem with considering correlations in the analysis of multiagent system with stochastic packet loss is that they induce dependencies between agents that are otherwise decoupled, preventing the application of decomposition methods required for efficient evaluation. To circumvent that issue, this paper is proposing an approach based on analysing sets of networks with independent communication links, only considering the correlations in an implicit fashion. Combining ideas from the robust stabilization of Markov jump linear systems with recently proposed techniques for analysing packet loss in multiagent systems, we obtain a linear matrix inequality based stability condition which is independent of the number of agents. The main result is that the set of stabilized probability distributions has non-empty interior such that small correlations cannot lead to instability, even though only distributions of independent links were analysed. Moreover, two examples are provided to demonstrate the applicability of the results to practically relevant scenarios.
comment: 7 pages, 2 figures, 1 table
☆ Cyclic pursuit formation control for arbitrary desired shapes
A multi-agent system comprises numerous agents that autonomously make decisions to collectively accomplish tasks, drawing significant attention for their wide-ranging applications. Within this context, formation control emerges as a prominent task, wherein agents collaboratively shape and maneuver while preserving formation integrity. Our focus centers on cyclic pursuit, a method facilitating the formation of circles, ellipses, and figure-eights under the assumption that agents can only perceive the relative positions of those preceding them. However, this method's scope has been restricted to these specific shapes, leaving the feasibility of forming other shapes uncertain. In response, our study proposes a novel method based on cyclic pursuit capable of forming a broader array of shapes, enabling agents to individually shape while pursuing preceding agents, thereby extending the repertoire of achievable formations. We present two scenarios concerning the information available to agents and devise formation control methods tailored to each scenario. Through extensive simulations, we demonstrate the efficacy of our proposed method in forming multiple shapes, including those represented as Fourier series, thereby underscoring the versatility and effectiveness of our approach.
☆ Natural-artificial hybrid swarm: Cyborg-insect group navigation in unknown obstructed soft terrain
Navigating multi-robot systems in complex terrains has always been a challenging task. This is due to the inherent limitations of traditional robots in collision avoidance, adaptation to unknown environments, and sustained energy efficiency. In order to overcome these limitations, this research proposes a solution by integrating living insects with miniature electronic controllers to enable robotic-like programmable control, and proposing a novel control algorithm for swarming. Although these creatures, called cyborg insects, have the ability to instinctively avoid collisions with neighbors and obstacles while adapting to complex terrains, there is a lack of literature on the control of multi-cyborg systems. This research gap is due to the difficulty in coordinating the movements of a cyborg system under the presence of insects' inherent individual variability in their reactions to control input. In response to this issue, we propose a novel swarm navigation algorithm addressing these challenges. The effectiveness of the algorithm is demonstrated through an experimental validation in which a cyborg swarm was successfully navigated through an unknown sandy field with obstacles and hills. This research contributes to the domain of swarm robotics and showcases the potential of integrating biological organisms with robotics and control theory to create more intelligent autonomous systems with real-world applications.
☆ Reverse Kron reduction of Multi-phase Radial Network
We consider the problem of identifying the admittance matrix of a three-phase radial network from voltage and current measurements at a subset of nodes. These measurements are used to estimate a virtual network represented by the Kron reduction (Schur complement) of the full admittance matrix. We focus on recovering exactly the full admittance matrix from its Kron reduction, i.e., computing the inverse of Schur complement. The key idea is to decompose Kron reduction into a sequence of iterations that maintains an invariance structure, and exploit this structure to reverse each step of the iterative Kron reduction.
☆ A Moreau Envelope Approach for LQR Meta-Policy Estimation
We study the problem of policy estimation for the Linear Quadratic Regulator (LQR) in discrete-time linear time-invariant uncertain dynamical systems. We propose a Moreau Envelope-based surrogate LQR cost, built from a finite set of realizations of the uncertain system, to define a meta-policy efficiently adjustable to new realizations. Moreover, we design an algorithm to find an approximate first-order stationary point of the meta-LQR cost function. Numerical results show that the proposed approach outperforms naive averaging of controllers on new realizations of the linear system. We also provide empirical evidence that our method has better sample complexity than Model-Agnostic Meta-Learning (MAML) approaches.
comment: 8 pages
Reinforcement Learning-based Receding Horizon Control using Adaptive Control Barrier Functions for Safety-Critical Systems
Optimal control methods provide solutions to safety-critical problems but easily become intractable. Control Barrier Functions (CBFs) have emerged as a popular technique that facilitates their solution by provably guaranteeing safety, through their forward invariance property, at the expense of some performance loss. This approach involves defining a performance objective alongside CBF-based safety constraints that must always be enforced. Unfortunately, both performance and solution feasibility can be significantly impacted by two key factors: (i) the selection of the cost function and associated parameters, and (ii) the calibration of parameters within the CBF-based constraints, which capture the trade-off between performance and conservativeness. %as well as infeasibility. To address these challenges, we propose a Reinforcement Learning (RL)-based Receding Horizon Control (RHC) approach leveraging Model Predictive Control (MPC) with CBFs (MPC-CBF). In particular, we parameterize our controller and use bilevel optimization, where RL is used to learn the optimal parameters while MPC computes the optimal control input. We validate our method by applying it to the challenging automated merging control problem for Connected and Automated Vehicles (CAVs) at conflicting roadways. Results demonstrate improved performance and a significant reduction in the number of infeasible cases compared to traditional heuristic approaches used for tuning CBF-based controllers, showcasing the effectiveness of the proposed method.
Destination-Constrained Linear Dynamical System Modeling in Set-Valued Frameworks
Directional motion towards a specified destination is a common occurrence in physical processes and human societal activities. Utilizing this prior information can significantly improve the control and predictive performance of system models. This paper primarily focuses on reconstructing linear dynamic system models based on destination constraints in the set-valued framework. We treat destination constraints as inherent information in the state evolution process and employ convex optimization techniques to construct a coherent and robust state model. This refined model effectively captures the impact of destination constraints on the state evolution at each time step. Furthermore, we design an optimal weight matrix for the reconstructed model to ensure smoother and more natural trajectories of state evolution. We also analyze the theoretical guarantee of optimality for this weight matrix and the properties of the reconstructed model. Finally, simulation experiments verify that the reconstructed model has significant advantages over the unconstrained and unoptimized weighted models and constrains the evolution of state trajectories with different starting and ending points.
comment: 15 pages, 11 figures
☆ Code Generation for Conic Model-Predictive Control on Microcontrollers with TinyMPC
Conic constraints appear in many important control applications like legged locomotion, robotic manipulation, and autonomous rocket landing. However, current solvers for conic optimization problems have relatively heavy computational demands in terms of both floating-point operations and memory footprint, making them impractical for use on small embedded devices. We extend TinyMPC, an open-source, high-speed solver targeting low-power embedded control applications, to handle second-order cone constraints. We also present code-generation software to enable deployment of TinyMPC on a variety of microcontrollers. We benchmark our generated code against state-of-the-art embedded QP and SOCP solvers, demonstrating a two-order-of-magnitude speed increase over ECOS while consuming less memory. Finally, we demonstrate TinyMPC's efficacy on the Crazyflie, a lightweight, resource-constrained quadrotor with fast dynamics. TinyMPC and its code-generation tools are publicly available at https://tinympc.org.
comment: Submitted to CDC, 2024. First two authors contributed equally
☆ Multiple Model Reference Adaptive Control with Blending for Non-Square Multivariable Systems
In this paper we develop a multiple model reference adaptive controller (MMRAC) with blending. The systems under consideration are non-square, i.e., the number of inputs is not equal to the number of states; multi-input, linear, time-invariant with uncertain parameters that lie inside of a known, compact, and convex set. Moreover, the full state of the plant is available for feedback. A multiple model online identification scheme for the plant's state and input matrices is developed that guarantees the estimated parameters converge to the underlying plant model under the assumption of persistence of excitation. Using an exact matching condition, the parameter estimates are used in a control law such that the plant's states asymptotically track the reference signal generated by a state-space model reference. The control architecture is proven to provide boundedness of all closed-loop signals and to asymptotically drive the state tracking error to zero. Numerical simulations illustrate the stability and efficacy of the proposed MMRAC scheme.
comment: 10 pages, 7 figures, IEEE Journal Submission
☆ ANOCA: AC Network-aware Optimal Curtailment Approach for Dynamic Hosting Capacity
With exponential growth in distributed energy resources (DERs) coupled with at-capacity distribution grid infrastructure, prosumers cannot always export all extra power to the grid without violating technical limits. Consequently, a slew of dynamic hosting capacity (DHC) algorithms have emerged for optimal utilization of grid infrastructure while maximizing export from DERs. Most of these DHC algorithms utilize the concept of operating envelopes (OE)}, where the utility gives prosumers technical power export limits, and they are free to export power within these limits. Recent studies have shown that OE-based frameworks have drawbacks, as most develop power export limits based on convex or linear grid models. As OEs must capture extreme operating conditions, both convex and linear models can violate technical limits in practice because they approximate grid physics. However, AC models are unsuitable because they may not be feasible within the whole region of OE. We propose a new two-stage optimization framework for DHC built on three-phase AC models to address the current gaps. In this approach, the prosumers first run a receding horizon multi-period optimization to identify optimal export power setpoints to communicate with the utility. The utility then performs an infeasibility-based optimization to either accept the prosumer's request or dispatch an optimal curtail signal such that overall system technical constraints are not violated. To explore various curtailment strategies, we develop an L1, L2, and Linf norm-based dispatch algorithm with an exact three-phase AC model. We test our framework on a 1420 three-phase node meshed distribution network and show that the proposed algorithm optimally curtails DERs while guaranteeing the AC feasibility of the network.
☆ From Sontag s to Cardano-Lyapunov Formula for Systems Not Affine in the Control: Convection-Enabled PDE Stabilization
We propose the first generalization of Sontag s universal controller to systems not affine in the control, particularly, to PDEs with boundary actuation. We assume that the system admits a control Lyapunov function (CLF) whose derivative, rather than being affine in the control, has either a depressed cubic, quadratic, or depressed quartic dependence on the control. For each case, a continuous universal controller that vanishes at the origin and achieves global exponential stability is derived. We prove our result in the context of convectionreaction-diffusion PDEs with Dirichlet actuation. We show that if the convection has a certain structure, then the L2 norm of the state is a CLF. In addition to generalizing Sontag s formula to some non-affine systems, we present the first general Lyapunov approach for boundary control of nonlinear PDEs. We illustrate our results via a numerical example.
comment: To be presented at the 2024 American Control Conference
☆ Path Integral Control with Rollout Clustering and Dynamic Obstacles
Model Predictive Path Integral (MPPI) control has proven to be a powerful tool for the control of uncertain systems (such as systems subject to disturbances and systems with unmodeled dynamics). One important limitation of the baseline MPPI algorithm is that it does not utilize simulated trajectories to their fullest extent. For one, it assumes that the average of all trajectories weighted by their performance index will be a safe trajectory. In this paper, multiple examples are shown where the previous assumption does not hold, and a trajectory clustering technique is presented that reduces the chances of the weighted average crossing in an unsafe region. Secondly, MPPI does not account for dynamic obstacles, so the authors put forward a novel cost function that accounts for dynamic obstacles without adding significant computation time to the overall algorithm. The novel contributions proposed in this paper were evaluated with extensive simulations to demonstrate improvements upon the state-of-the-art MPPI techniques.
comment: 8 pages, 5 figures, extended version of ACC 2024 submission
☆ Adaptive Boundary Control of the Kuramoto-Sivashinsky Equation Under Intermittent Sensing
We study in this paper boundary stabilization, in the L2 sense, of the one-dimensional Kuramoto-Sivashinsky equation subject to intermittent sensing. We assume that we measure the state of this spatio-temporal equation on a given spatial subdomain during certain intervals of time, while we measure the state on the remaining spatial subdomain during the remaining intervals of time. As a result, we assign a feedback law at the boundary of the spatial domain and force to zero the value of the state at the junction of the two subdomains. Throughout the study, the destabilizing coefficient is assumed to be space-dependent and bounded but unknown. Adaptive boundary controllers are designed under different assumptions on the forcing term. In particular, when the forcing term is null, we guarantee global exponential stability of the origin. Furthermore, when the forcing term is bounded and admits a known upper bound, we guarantee input-to-state stability, and only global uniform ultimate boundedness is guaranteed when the upper bound is unknown. Numerical simulations are performed to illustrate our results
comment: Submitted to Automatica
☆ Learning Piecewise Residuals of Control Barrier Functions for Safety of Switching Systems using Multi-Output Gaussian Processes
Control barrier functions (CBFs) have recently been introduced as a systematic tool to ensure safety by establishing set invariance. When combined with a control Lyapunov function (CLF), they form a safety-critical control mechanism. However, the effectiveness of CBFs and CLFs is closely tied to the system model. In practice, model uncertainty can jeopardize safety and stability guarantees and may lead to undesirable performance. In this paper, we develop a safe learning-based control strategy for switching systems in the face of uncertainty. We focus on the case that a nominal model is available for a true underlying switching system. This uncertainty results in piecewise residuals for each switching surface, impacting the CLF and CBF constraints. We introduce a batch multi-output Gaussian process (MOGP) framework to approximate these piecewise residuals, thereby mitigating the adverse effects of uncertainty. A particular structure of the covariance function enables us to convert the MOGP-based chance constraints CLF and CBF into second-order cone constraints, which leads to a convex optimization. We analyze the feasibility of the resulting optimization and provide the necessary and sufficient conditions for feasibility. The effectiveness of the proposed strategy is validated through a simulation of a switching adaptive cruise control system.
comment: arXiv admin note: text overlap with arXiv:2403.09573
☆ A Constructive Method for Designing Safe Multirate Controllers for Differentially-Flat Systems
We present a multi-rate control architecture that leverages fundamental properties of differential flatness to synthesize controllers for safety-critical nonlinear dynamical systems. We propose a two-layer architecture, where the high-level generates reference trajectories using a linear Model Predictive Controller, and the low-level tracks this reference using a feedback controller. The novelty lies in how we couple these layers, to achieve formal guarantees on recursive feasibility of the MPC problem, and safety of the nonlinear system. Furthermore, using differential flatness, we provide a constructive means to synthesize the multi-rate controller, thereby removing the need to search for suitable Lyapunov or barrier functions, or to approximately linearize/discretize nonlinear dynamics. We show the synthesized controller is a convex optimization problem, making it amenable to real-time implementations. The method is demonstrated experimentally on a ground rover and a quadruped robotic system.
comment: 6 pages, 3 figures, accepted at IEEE Control Systems Letters 2021
♻ ☆ Optimal Control of Grid-Interfacing Inverters With Current Magnitude Limits
Grid-interfacing inverters act as the interface between renewable resources and the electric grid, and have the potential to offer fast and programmable controls compared to synchronous generators. With this flexibility there has been significant research efforts into determining the best way to control these inverters. Inverters are limited in their maximum current output in order to protect semiconductor devices, presenting a nonlinear constraint that needs to be accounted for in their control algorithms. Existing approaches either simply saturate a controller that is designed for unconstrained systems, or assume small perturbations and linearize a saturated system. These approaches can lead to stability issues or limiting the control actions to be too conservative. In this paper, we directly focus on a nonlinear system that explicitly accounts for the saturation of the current magnitude. We use a Lyapunov stability approach to determine a stability condition for the system, guaranteeing that a class of controllers would be stabilizing if they satisfy a simple SDP condition. With this condition we fit a linear-feedback controller by sampling the output (offline) model predictive control problems. This learned controller has improved performances with existing designs.
comment: 6 pages, 6 figures, 1 table. Submitted to CDC'2024
♻ ☆ Faster Run-to-Run Feedforward Control of Electromechanical Switching Devices: a Sensitivity-Based Approach
Electromechanical switching devices, such as solenoid valves, contactors, and relays, suffer from undesirable phenomena like clicking, mechanical wear, and contact bounce. Despite that, they are still widely used in industry due to their various economic and technical advantages. This has encouraged the development of controllers aimed at reducing the collisions that occur at the end of the switching operations. One of the most successful approaches has been the use of iterative techniques. However, these algorithms typically require a large number of operations to converge, which is definitely a clear drawback. This paper presents a strategy to improve the convergence rate of such controllers. Our proposal, which is based on the sensitivity of the control law with respect to the parameters, assumes that the performance of the system is more heavily affected by some parameters than others. Thus, by avoiding movements in the directions that have less impact, the search algorithm is expected to drive the system to near-optimal behaviors using fewer operations. Results obtained by simulation show significant improvement in the convergence rate of a state-of-the-art run-to-run feedforward controller, which demonstrates the high potential of the proposal.
comment: 6 pages, 4 figures. Final version, after peer review and acceptance, submitted to the 22nd European Control Conference (ECC)
♻ ☆ Resilient source seeking with robot swarms
We present a solution for locating the source, or maximum, of an unknown scalar field using a swarm of mobile robots. Unlike relying on the traditional gradient information, the swarm determines an ascending direction to approach the source with arbitrary precision. The ascending direction is calculated from measurements of the field strength at the robot locations and their relative positions concerning the centroid. Rather than focusing on individual robots, we focus the analysis on the density of robots per unit area to guarantee a more resilient swarm, i.e., the functionality remains even if individuals go missing or are misplaced during the mission. We reinforce the robustness of the algorithm by providing sufficient conditions for the swarm shape so that the ascending direction is almost parallel to the gradient. The swarm can respond to an unexpected environment by morphing its shape and exploiting the existence of multiple ascending directions. Finally, we validate our approach numerically with hundreds of robots. The fact that a large number of robots always calculate an ascending direction compensates for the loss of individuals and mitigates issues arising from the actuator and sensor noises.
comment: 7 pages, submitted to CDC 2024
♻ ☆ Frequency Regulation with Storage: On Losses and Profits
Low-carbon societies will need to store vast amounts of electricity to balance intermittent generation from wind and solar energy, for example, through frequency regulation. Here, we derive an analytical solution to the decision-making problem of storage operators who sell frequency regulation power to grid operators and trade electricity on day-ahead markets. Mathematically, we treat future frequency deviation trajectories as functional uncertainties in a receding horizon robust optimization problem. We constrain the expected terminal state-of-charge to be equal to some target to allow storage operators to make good decisions not only for the present but also the future. Thanks to this constraint, the amount of electricity traded on day-ahead markets is an implicit function of the regulation power sold to grid operators. The implicit function quantifies the amount of power that needs to be purchased to cover the expected energy loss that results from providing frequency regulation. We show how the marginal cost associated with the expected energy loss decreases with roundtrip efficiency and increases with frequency deviation dispersion. We find that the profits from frequency regulation over the lifetime of energy-constrained storage devices are roughly inversely proportional to the length of time for which regulation power must be committed.
♻ ☆ A Safe Preference Learning Approach for Personalization with Applications to Autonomous Vehicles
This work introduces a preference learning method that ensures adherence to given specifications, with an application to autonomous vehicles. Our approach incorporates the priority ordering of Signal Temporal Logic (STL) formulas describing traffic rules into a learning framework. By leveraging Parametric Weighted Signal Temporal Logic (PWSTL), we formulate the problem of safety-guaranteed preference learning based on pairwise comparisons and propose an approach to solve this learning problem. Our approach finds a feasible valuation for the weights of the given PWSTL formula such that, with these weights, preferred signals have weighted quantitative satisfaction measures greater than their non-preferred counterparts. The feasible valuation of weights given by our approach leads to a weighted STL formula that can be used in correct-and-custom-by-construction controller synthesis. We demonstrate the performance of our method with a pilot human subject study in two different simulated driving scenarios involving a stop sign and a pedestrian crossing. Our approach yields competitive results compared to existing preference learning methods in terms of capturing preferences and notably outperforms them when safety is considered.
comment: 9 pages, 3 figures, 2 tables. This work has been published at IEEE Robotics and Automation Letters. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Ensuring Disturbance Rejection Performance by Synthesizing Grid-Following and Grid-Forming Inverters in Power Systems
To satisfy dynamic requirements of power systems, it is imperative for grid-tied inverters to ensure good disturbance rejection performance (DRP) under variable grid conditions. This letter discovers and theoretically proves that for general networks, synthesizing grid-following (GFL) inverters and grid-forming (GFM) inverters can always more effectively ensure the DRP of multiple inverters, as compared to homogeneous inverter-based systems that solely utilize either GFL or GFM inverters. The synthesis of GFL inverters and GFM inverters can concurrently increase the smallest eigenvalue and decrease the largest eigenvalue of the network grounded Laplacian matrix. This can be equivalent to rematching the proper short-circuit ratio (SCR) for GFL and GFM inverters, thereby ensuring the DRP of inverters both in weak and strong grids. The results reveal the necessity of synthesizing diverse inverter control schemes from the network-based perspective. Sensitivity function-based tests and real-time simulations validate our results.
comment: 5 pages
♻ ☆ Closing the Gap to Quadratic Invariance: a Regret Minimization Approach to Optimal Distributed Control
In this work, we focus on the design of optimal controllers that must comply with an information structure. State-of-the-art approaches do so based on the H2 or Hinfty norm to minimize the expected or worst-case cost in the presence of stochastic or adversarial disturbances. Large-scale systems often experience a combination of stochastic and deterministic disruptions (e.g., sensor failures, environmental fluctuations) that spread across the system and are difficult to model precisely, leading to sub-optimal closed-loop behaviors. Hence, we propose improving performance for these scenarios by minimizing the regret with respect to an ideal policy that complies with less stringent sensor-information constraints. This endows our controller with the ability to approach the improved behavior of a more informed policy, which would detect and counteract heterogeneous and localized disturbances more promptly. Specifically, we derive convex relaxations of the resulting regret minimization problem that are compatible with any desired controller sparsity, while we reveal a renewed role of the Quadratic Invariance (QI) condition in designing informative benchmarks to measure regret. Last, we validate our proposed method through numerical simulations on controlling a multi-agent distributed system, comparing its performance with traditional H2 and Hinfty policies.
comment: Accepted for presentation and publication in the proceedings of the 2024 European Control Conference (ECC)
♻ ☆ Time-Optimal Control for High-Order Chain-of-Integrators Systems with Full State Constraints and Arbitrary Terminal States (Extended Version)
Time-optimal control for high-order chain-of-integrators systems with full state constraints and arbitrary given terminal states remains a challenging problem in the optimal control theory domain, yet to be resolved. To enhance further comprehension of the problem, this paper establishes a novel notation system and theoretical framework, providing the switching manifold for high-order problems in the form of switching laws. Through deriving properties of switching laws on signs and dimension, this paper proposes a definite condition for time-optimal control. Guided by the developed theory, a trajectory planning method named the manifold-intercept method (MIM) is developed. The proposed MIM can plan time-optimal jerk-limited trajectories with full state constraints, and can also plan near-optimal non-chattering higher-order trajectories with negligible extra motion time compared to optimal profiles. Numerical results indicate that the proposed MIM outperforms all baselines in computational time, computational accuracy, and trajectory quality by a large gap.
♻ ☆ Discretized Distributed Optimization over Dynamic Digraphs
We consider a discrete-time model of continuous-time distributed optimization over dynamic directed-graphs (digraphs) with applications to distributed learning. Our optimization algorithm works over general strongly connected dynamic networks under switching topologies, e.g., in mobile multi-agent systems and volatile networks due to link failures. Compared to many existing lines of work, there is no need for bi-stochastic weight designs on the links. The existing literature mostly needs the link weights to be stochastic using specific weight-design algorithms needed both at the initialization and at all times when the topology of the network changes. This paper eliminates the need for such algorithms and paves the way for distributed optimization over time-varying digraphs. We derive the bound on the gradient-tracking step-size and discrete time-step for convergence and prove dynamic stability using arguments from consensus algorithms, matrix perturbation theory, and Lyapunov theory. This work, particularly, is an improvement over existing stochastic-weight undirected networks in case of link removal or packet drops. This is because the existing literature may need to rerun time-consuming and computationally complex algorithms for stochastic design, while the proposed strategy works as long as the underlying network is weight-symmetric and balanced. The proposed optimization framework finds applications to distributed classification and learning.
♻ ☆ When Robotics Meets Wireless Communications: An Introductory Tutorial
The importance of ground Mobile Robots (MRs) and Unmanned Aerial Vehicles (UAVs) within the research community, industry, and society is growing fast. Many of these agents are nowadays equipped with communication systems that are, in some cases, essential to successfully achieve certain tasks. In this context, we have begun to witness the development of a new interdisciplinary research field at the intersection of robotics and communications. This research field has been boosted by the intention of integrating UAVs within the 5G and 6G communication networks. This research will undoubtedly lead to many important applications in the near future. Nevertheless, one of the main obstacles to the development of this research area is that most researchers address these problems by oversimplifying either the robotics or the communications aspect. This impedes the ability of reaching the full potential of this new interdisciplinary research area. In this tutorial, we present some of the modelling tools necessary to address problems involving both robotics and communication from an interdisciplinary perspective. As an illustrative example of such problems, we focus in this tutorial on the issue of communication-aware trajectory planning.
comment: 35 pages, 192 references
♻ ☆ Stable Linear Subspace Identification: A Machine Learning Approach
Machine Learning (ML) and linear System Identification (SI) have been historically developed independently. In this paper, we leverage well-established ML tools - especially the automatic differentiation framework - to introduce SIMBa, a family of discrete linear multi-step-ahead state-space SI methods using backpropagation. SIMBa relies on a novel Linear-Matrix-Inequality-based free parametrization of Schur matrices to ensure the stability of the identified model. We show how SIMBa generally outperforms traditional linear state-space SI methods, and sometimes significantly, although at the price of a higher computational burden. This performance gap is particularly remarkable compared to other SI methods with stability guarantees, where the gain is frequently above 25% in our investigations, hinting at SIMBa's ability to simultaneously achieve state-of-the-art fitting performance and enforce stability. Interestingly, these observations hold for a wide variety of input-output systems and on both simulated and real-world data, showcasing the flexibility of the proposed approach. We postulate that this new SI paradigm presents a great extension potential to identify structured nonlinear models from data, and we hence open-source SIMBa on https://github.com/Cemempamoi/simba.
comment: Accepted at ECC 2024
♻ ☆ Modelling Irrational Behaviour of Residential End Users using Non-Stationary Gaussian Processes
Demand response (DR) plays a critical role in ensuring efficient electricity consumption and optimal use of network assets. Yet, existing DR models often overlook a crucial element, the irrational behaviour of electricity end users. In this work, we propose a price-responsive model that incorporates key aspects of end-user irrationality, specifically loss aversion, time inconsistency, and bounded rationality. To this end, we first develop a framework that uses Multiple Seasonal-Trend decomposition using Loess (MSTL) and non-stationary Gaussian processes to model the randomness in the electricity consumption by residential consumers. The impact of this model is then evaluated through a community battery storage (CBS) business model. Additionally, we apply a chance-constrained optimisation model for CBS operation that deals with the unpredictability of the end-user irrationality. Our simulations using real-world data show that the proposed DR model provides a more realistic estimate of end-user price-responsive behaviour when considering irrationality. Compared to a deterministic model that cannot fully take into account the irrational behaviour of end users, the chance-constrained CBS operation model yields an additional 19% revenue. Lastly, the business model reduces the electricity costs of solar end users by 11%.
comment: This manuscript has been accepted for publication in IEEE Transactions on Smart Grid
♻ ☆ Full Attitude Intelligent Controller Design of a Heliquad under Complete Failure of an Actuator
In this paper, we design a reliable Heliquad and develop an intelligent controller to handle one actuators complete failure. Heliquad is a multi-copter similar to Quadcopter, with four actuators diagonally symmetric from the center. Each actuator has two control inputs; the first input changes the propeller blades collective pitch (also called variable pitch), and the other input changes the rotation speed. For reliable operation and high torque characteristic requirement for yaw control, a cambered airfoil is used to design propeller blades. A neural network-based control allocation is designed to provide complete control authority even under a complete loss of one actuator. Nonlinear quaternion based outer loop position control, with proportional-derivative inner loop for attitude control and neural network-based control allocation is used in controller design. The proposed controller and Heliquad designs performance is evaluated using a software-in-loop simulation to track the position reference command under failure. The results clearly indicate that the Heliquad with an intelligent controller provides necessary tracking performance even under a complete loss of one actuator.
comment: 7 pages, For video go to https://indianinstituteofscience-my.sharepoint.com/:v:/g/personal/eeshank_iisc_ac_in/EcMg2uTtE91AsHDejNkb6YMBNckaXGjeh_YMzDV6sAHZAQ?e=DrRqmN
♻ ☆ Autonomous Hook-Based Grasping and Transportation with Quadcopters
Payload grasping and transportation with quadcopters is an active research area that has rapidly developed over the last decade. To grasp a payload without human interaction, most state-of-the-art approaches apply robotic arms that are attached to the quadcopter body. However, due to the large weight and power consumption of these aerial manipulators, their agility and flight time are limited. This paper proposes a motion control and planning method for transportation with a lightweight, passive manipulator structure that consists of a hook attached to a quadrotor using a 1 DoF revolute joint. To perform payload grasping, transportation, and release, first, time-optimal reference trajectories are designed through specific waypoints to ensure the fast and reliable execution of the tasks. Then, a two-stage motion control approach is developed based on a robust geometric controller for precise and reliable reference tracking and a linear--quadratic payload regulator for rapid setpoint stabilization of the payload swing. Furthermore, stability of the closed-loop system is mathematically proven to give safety guarantee for its operation. The proposed control architecture and design are evaluated in a high-fidelity physical simulator, and also in real flight experiments, using a custom-made quadrotor--hook manipulator platform.
♻ ☆ Robustness Evaluation of Localization Techniques for Autonomous Racing
This work introduces SynPF, an MCL-based algorithm tailored for high-speed racing environments. Benchmarked against Cartographer, a state-of-the-art pose-graph SLAM algorithm, SynPF leverages synergies from previous particle-filtering methods and synthesizes them for the high-performance racing domain. Our extensive in-field evaluations reveal that while Cartographer excels under nominal conditions, it struggles when subjected to wheel-slip, a common phenomenon in a racing scenario due to varying grip levels and aggressive driving behaviour. Conversely, SynPF demonstrates robustness in these challenging conditions and a low-latency computation time of 1.25 ms on on-board computers without a GPU. Using the F1TENTH platform, a 1:10 scaled autonomous racing vehicle, this work not only highlights the vulnerabilities of existing algorithms in high-speed scenarios, tested up until 7.6 m/s, but also emphasizes the potential of SynPF as a viable alternative, especially in deteriorating odometry conditions.
comment: Accepted at the Design, Automation and Test in Europe Conference 2024 as an extended abstract
♻ ☆ Credit vs. Discount-Based Congestion Pricing: A Comparison Study
Tolling, or congestion pricing, offers a promising traffic management policy for regulating congestion, but has also attracted criticism for placing outsized financial burdens on low-income users. Credit-based congestion pricing (CBCP) and discount-based congestion pricing (DBCP) policies, which respectively provide travel credits and toll discounts to low-income users on tolled roads, have emerged as promising mechanisms for reducing traffic congestion without worsening societal inequities. However, the optimal design of CBCP and DBCP policies, as well as their relative advantages and disadvantages, remain poorly understood. To address this, we study the effects of implementing CBCP and DBCP policies to route users on a network of multi-lane highways with tolled express lanes. We formulate a non-atomic routing game framework in which a subset of eligible users is granted toll relief in the form of a fixed budget or toll discount, while the remaining ineligible users must pay out-of-pocket. We prove the existence of Nash equilibrium traffic flow patterns corresponding to any given CBCP or DBCP policy. Under the additional assumption that eligible users have time-invariant VoTs, we provide a convex program to efficiently compute these equilibria. For networks consisting of a single edge, we identify conditions under which CBCP policies outperform DBCP policies (and vice versa), in the sense of improving eligible users' access to the express lane. Finally, we present empirical results from a CBCP pilot study of the San Mateo 101 Express Lane Project in California. Our empirical results corroborate our theoretical analysis of the impact of deploying credit-based and discount-based policies, and lend insights into the sensitivity of their impact with respect to the travel demand and users' VoTs.
♻ ☆ Achievable Sum Rate Optimization on NOMA-aided Cell-Free Massive MIMO with Finite Blocklength Coding
Non-orthogonal multiple access (NOMA)-aided cell-free massive multiple-input multiple-output (CFmMIMO) has been considered as a promising technology to fulfill strict quality of service requirements for ultra-reliable low-latency communications (URLLC). However, finite blocklength coding (FBC) in URLLC makes it challenging to achieve the optimal performance in the NOMA-aided CFmMIMO system. In this paper, we investigate the performance of the NOMA-aided CFmMIMO system with FBC in terms of achievable sum rate (ASR). Firstly, we derive a lower bound (LB) on the ergodic data rate. Then, we formulate an ASR maximization problem by jointly considering power allocation and user equipment (UE) clustering. To tackle such an intractable problem, we decompose it into two sub-problems, i.e., the power allocation problem and the UE clustering problem. A successive convex approximation (SCA) algorithm is proposed to solve the power allocation problem by transforming it into a series of geometric programming problems. Meanwhile, two algorithms based on graph theory are proposed to solve the UE clustering problem by identifying negative loops. Finally, alternative optimization is performed to find the maximum ASR of the NOMA-aided CFmMIMO system with FBC. The simulation results demonstrate that the proposed algorithms significantly outperform the benchmark algorithms in terms of ASR under various scenarios.
♻ ☆ Compositional Inductive Invariant Based Verification of Neural Network Controlled Systems
The integration of neural networks into safety-critical systems has shown great potential in recent years. However, the challenge of effectively verifying the safety of Neural Network Controlled Systems (NNCS) persists. This paper introduces a novel approach to NNCS safety verification, leveraging the inductive invariant method. Verifying the inductiveness of a candidate inductive invariant in the context of NNCS is hard because of the scale and nonlinearity of neural networks. Our compositional method makes this verification process manageable by decomposing the inductiveness proof obligation into smaller, more tractable subproblems. Alongside the high-level method, we present an algorithm capable of automatically verifying the inductiveness of given candidates by automatically inferring the necessary decomposition predicates. The algorithm significantly outperforms the baseline method and shows remarkable reductions in execution time in our case studies, shortening the verification time from hours (or timeout) to seconds.
Robotics 85
☆ A Robotic Skill Learning System Built Upon Diffusion Policies and Foundation Models
In this paper, we build upon two major recent developments in the field, Diffusion Policies for visuomotor manipulation and large pre-trained multimodal foundational models to obtain a robotic skill learning system. The system can obtain new skills via the behavioral cloning approach of visuomotor diffusion policies given teleoperated demonstrations. Foundational models are being used to perform skill selection given the user's prompt in natural language. Before executing a skill the foundational model performs a precondition check given an observation of the workspace. We compare the performance of different foundational models to this end as well as give a detailed experimental evaluation of the skills taught by the user in simulation and the real world. Finally, we showcase the combined system on a challenging food serving scenario in the real world. Videos of all experimental executions, as well as the process of teaching new skills in simulation and the real world, are available on the project's website.
comment: https://roboskillframework.github.io
☆ BatDeck: Advancing Nano-drone Navigation with Low-power Ultrasound-based Obstacle Avoidance
Nano-drones, distinguished by their agility, minimal weight, and cost-effectiveness, are particularly well-suited for exploration in confined, cluttered and narrow spaces. Recognizing transparent, highly reflective or absorbing materials, such as glass and metallic surfaces is challenging, as classical sensors, such as cameras or laser rangers, often do not detect them. Inspired by bats, which can fly at high speeds in complete darkness with the help of ultrasound, this paper introduces \textit{BatDeck}, a pioneering sensor-deck employing a lightweight and low-power ultrasonic sensor for nano-drone autonomous navigation. This paper first provides insights about sensor characteristics, highlighting the influence of motor noise on the ultrasound readings, then it introduces the results of extensive experimental tests for obstacle avoidance (OA) in a diverse environment. Results show that \textit{BatDeck} allows exploration for a flight time of 8 minutes while covering 136m on average before crash in a challenging environment with transparent and reflective obstacles, proving the effectiveness of ultrasonic sensors for OA on nano-drones.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
☆ Synapse: Learning Preferential Concepts from Visual Demonstrations
This paper addresses the problem of preference learning, which aims to learn user-specific preferences (e.g., "good parking spot", "convenient drop-off location") from visual input. Despite its similarity to learning factual concepts (e.g., "red cube"), preference learning is a fundamentally harder problem due to its subjective nature and the paucity of person-specific training data. We address this problem using a new framework called Synapse, which is a neuro-symbolic approach designed to efficiently learn preferential concepts from limited demonstrations. Synapse represents preferences as neuro-symbolic programs in a domain-specific language (DSL) that operates over images, and leverages a novel combination of visual parsing, large language models, and program synthesis to learn programs representing individual preferences. We evaluate Synapse through extensive experimentation including a user case study focusing on mobility-related concepts in mobile robotics and autonomous driving. Our evaluation demonstrates that Synapse significantly outperforms existing baselines as well as its own ablations. The code and other details can be found on the project website https://amrl.cs.utexas.edu/synapse .
comment: 23 pages, 7 figures; Preprint
☆ Domain Adaptive Detection of MAVs: A Benchmark and Noise Suppression Network
Visual detection of Micro Air Vehicles (MAVs) has attracted increasing attention in recent years due to its important application in various tasks. The existing methods for MAV detection assume that the training set and testing set have the same distribution. As a result, when deployed in new domains, the detectors would have a significant performance degradation due to domain discrepancy. In this paper, we study the problem of cross-domain MAV detection. The contributions of this paper are threefold. 1) We propose a Multi-MAV-Multi-Domain (M3D) dataset consisting of both simulation and realistic images. Compared to other existing datasets, the proposed one is more comprehensive in the sense that it covers rich scenes, diverse MAV types, and various viewing angles. A new benchmark for cross-domain MAV detection is proposed based on the proposed dataset. 2) We propose a Noise Suppression Network (NSN) based on the framework of pseudo-labeling and a large-to-small training procedure. To reduce the challenging pseudo-label noises, two novel modules are designed in this network. The first is a prior-based curriculum learning module for allocating adaptive thresholds for pseudo labels with different difficulties. The second is a masked copy-paste augmentation module for pasting truly-labeled MAVs on unlabeled target images and thus decreasing pseudo-label noises. 3) Extensive experimental results verify the superior performance of the proposed method compared to the state-of-the-art ones. In particular, it achieves mAP of 46.9%(+5.8%), 50.5%(+3.7%), and 61.5%(+11.3%) on the tasks of simulation-to-real adaptation, cross-scene adaptation, and cross-camera adaptation, respectively.
comment: 17 pages, 11 figures. Accepted by IEEE Transactions on Automation Science and Engineering
☆ Skill Q-Network: Learning Adaptive Skill Ensemble for Mapless Navigation in Unknown Environments
This paper focuses on the acquisition of mapless navigation skills within unknown environments. We introduce the Skill Q-Network (SQN), a novel reinforcement learning method featuring an adaptive skill ensemble mechanism. Unlike existing methods, our model concurrently learns a high-level skill decision process alongside multiple low-level navigation skills, all without the need for prior knowledge. Leveraging a tailored reward function for mapless navigation, the SQN is capable of learning adaptive maneuvers that incorporate both exploration and goal-directed skills, enabling effective navigation in new environments. Our experiments demonstrate that our SQN can effectively navigate complex environments, exhibiting a 40% higher performance compared to baseline models. Without explicit guidance, SQN discovers how to combine low-level skill policies, showcasing both goal-directed navigations to reach destinations and exploration maneuvers to escape from local minimum regions in challenging scenarios. Remarkably, our adaptive skill ensemble method enables zero-shot transfer to out-of-distribution domains, characterized by unseen observations from non-convex obstacles or uneven, subterranean-like environments.
comment: 8 pages, 8 figures
Trajectory Planning of Robotic Manipulator in Dynamic Environment Exploiting DRL
This study is about the implementation of a reinforcement learning algorithm in the trajectory planning of manipulators. We have a 7-DOF robotic arm to pick and place the randomly placed block at a random target point in an unknown environment. The obstacle is randomly moving which creates a hurdle in picking the object. The objective of the robot is to avoid the obstacle and pick the block with constraints to a fixed timestamp. In this literature, we have applied a deep deterministic policy gradient (DDPG) algorithm and compared the model's efficiency with dense and sparse rewards.
comment: Accepted in ICIESTR-2024
☆ Bridging the Sim-to-Real Gap with Bayesian Inference
We present SIM-FSVGD for learning robot dynamics from data. As opposed to traditional methods, SIM-FSVGD leverages low-fidelity physical priors, e.g., in the form of simulators, to regularize the training of neural network models. While learning accurate dynamics already in the low data regime, SIM-FSVGD scales and excels also when more data is available. We empirically show that learning with implicit physical priors results in accurate mean model estimation as well as precise uncertainty quantification. We demonstrate the effectiveness of SIM-FSVGD in bridging the sim-to-real gap on a high-performance RC racecar system. Using model-based RL, we demonstrate a highly dynamic parking maneuver with drifting, using less than half the data compared to the state of the art.
☆ Symbolic and User-friendly Geometric Algebra Routines (SUGAR) for Computations in Matlab
Geometric algebra (GA) is a mathematical tool for geometric computing, providing a framework that allows a unified and compact approach to geometric relations which in other mathematical systems are typically described using different more complicated elements. This fact has led to an increasing adoption of GA in applied mathematics and engineering problems. However, the scarcity of symbolic implementations of GA and its inherent complexity, requiring a specific mathematical background, make it challenging and less intuitive for engineers to work with. This prevents wider adoption among more applied professionals. To address this challenge, this paper introduces SUGAR (Symbolic and User-friendly Geometric Algebra Routines), an open-source toolbox designed for Matlab and licensed under the MIT License. SUGAR facilitates the translation of GA concepts into Matlab and provides a collection of user-friendly functions tailored for GA computations, including support for symbolic operations. It supports both numeric and symbolic computations in high-dimensional GAs. Specifically tailored for applied mathematics and engineering applications, SUGAR has been meticulously engineered to represent geometric elements and transformations within two and three-dimensional projective and conformal geometric algebras, aligning with established computational methodologies in the literature. Furthermore, SUGAR efficiently handles functions of multivectors, such as exponential, logarithmic, sinusoidal, and cosine functions, enhancing its applicability across various engineering domains, including robotics, control systems, and power electronics. Finally, this work includes four distinct validation examples, demonstrating SUGAR's capabilities across the above-mentioned fields and its practical utility in addressing real-world applied mathematics and engineering problems.
comment: 33 pages, 6 figures, journal paper submitted to ACM TOMS
☆ Technical Development of a Semi-Autonomous Robotic Partition
This technical description details the design and engineering process of a semi-autonomous robotic partition. This robotic partition prototype was subsequently employed in a longer-term evaluation in-the-wild study conducted by the authors in a real-world office setting.
☆ ROXIE: Defining a Robotic eXplanation and Interpretability Engine IROS 2024
In an era where autonomous robots increasingly inhabit public spaces, the imperative for transparency and interpretability in their decision-making processes becomes paramount. This paper presents the overview of a Robotic eXplanation and Interpretability Engine (ROXIE), which addresses this critical need, aiming to demystify the opaque nature of complex robotic behaviors. This paper elucidates the key features and requirements needed for providing information and explanations about robot decision-making processes. It also overviews the suite of software components and libraries available for deployment with ROS 2, empowering users to provide comprehensive explanations and interpretations of robot processes and behaviors, thereby fostering trust and collaboration in human-robot interactions.
comment: 7 pages, 3 figures, 1 tables, Submitted to IROS 2024
☆ Research Challenges for Adaptive Architecture: Empowering Occupants of Multi-Occupancy Buildings
This positional paper outlines our vision of 'adaptive architecture', which involves the integration of robotic technology to physically change an architectural space in supporting the changing needs of its occupants, in response to the CHI'24 workshop "HabiTech - Inhabiting Buildings, Data & Technology" call on "How do new technologies enable and empower the inhabitants of multi-occupancy buildings?". Specifically, while adaptive architecture holds promise for enhancing occupant satisfaction, comfort, and overall health and well-being, there remains a range of research challenges of (1) how it can effectively support individual occupants, while (2) mediating the conflicting needs of collocated others, and (3) integrating meaningfully into the sociocultural characteristics of their building community.
☆ The Adaptive Workplace: Orchestrating Architectural Services around the Wellbeing of Individual Occupants
As the academic consortia members of the EU Horizon project SONATA ("Situation-aware OrchestratioN of AdapTive Architecture"), we respond to the workshop call for "Office Wellbeing by Design: Don't Stand for Anything Less" by proposing the "Adaptive Workplace" concept. In essence, our vision aims to adapt a workplace to the ever-changing needs of individual occupants, instead of that occupants are expected to adapt to their workplace.
☆ Counter-example guided Imitation Learning of Feedback Controllers from Temporal Logic Specifications
We present a novel method for imitation learning for control requirements expressed using Signal Temporal Logic (STL). More concretely we focus on the problem of training a neural network to imitate a complex controller. The learning process is guided by efficient data aggregation based on counter-examples and a coverage measure. Moreover, we introduce a method to evaluate the performance of the learned controller via parameterization and parameter estimation of the STL requirements. We demonstrate our approach with a flying robot case study.
☆ Active Admittance Control with Iterative Learning for General-Purpose Contact-Rich Manipulation
Force interaction is inevitable when robots face multiple operation scenarios. How to make the robot competent in force control for generalized operations such as multi-tasks still remains a challenging problem. Aiming at the reproducibility of interaction tasks and the lack of a generalized force control framework for multi-task scenarios, this paper proposes a novel hybrid control framework based on active admittance control with iterative learning parameters-tunning mechanism. The method adopts admittance control as the underlying algorithm to ensure flexibility, and iterative learning as the high-level algorithm to regulate the parameters of the admittance model. The whole algorithm has flexibility and learning ability, which is capable of achieving the goal of excellent versatility. Four representative interactive robot manipulation tasks are chosen to investigate the consistency and generalisability of the proposed method. Experiments are designed to verify the effectiveness of the whole framework, and an average of 98.21% and 91.52% improvement of RMSE is obtained relative to the traditional admittance control as well as the model-free adaptive control, respectively.
☆ Arm-Constrained Curriculum Learning for Loco-Manipulation of the Wheel-Legged Robot
Incorporating a robotic manipulator into a wheel-legged robot enhances its agility and expands its potential for practical applications. However, the presence of potential instability and uncertainties presents additional challenges for control objectives. In this paper, we introduce an arm-constrained curriculum learning architecture to tackle the issues introduced by adding the manipulator. Firstly, we develop an arm-constrained reinforcement learning algorithm to ensure safety and stability in control performance. Additionally, to address discrepancies in reward settings between the arm and the base, we propose a reward-aware curriculum learning method. The policy is first trained in Isaac gym and transferred to the physical robot to do dynamic grasping tasks, including the door-opening task, fan-twitching task and the relay-baton-picking and following task. The results demonstrate that our proposed approach effectively controls the arm-equipped wheel-legged robot to master dynamic grasping skills, allowing it to chase and catch a moving object while in motion. The code can be found at https://github.com/aCodeDog/legged-robots-manipulation. To view the supplemental video, please visit https://youtu.be/sNXT-rwPNMM.
☆ Hallucination Detection in Foundation Models for Decision-Making: A Flexible Definition and Review of the State of the Art
Autonomous systems are soon to be ubiquitous, from manufacturing autonomy to agricultural field robots, and from health care assistants to the entertainment industry. The majority of these systems are developed with modular sub-components for decision-making, planning, and control that may be hand-engineered or learning-based. While these existing approaches have been shown to perform well under the situations they were specifically designed for, they can perform especially poorly in rare, out-of-distribution scenarios that will undoubtedly arise at test-time. The rise of foundation models trained on multiple tasks with impressively large datasets from a variety of fields has led researchers to believe that these models may provide common sense reasoning that existing planners are missing. Researchers posit that this common sense reasoning will bridge the gap between algorithm development and deployment to out-of-distribution tasks, like how humans adapt to unexpected scenarios. Large language models have already penetrated the robotics and autonomous systems domains as researchers are scrambling to showcase their potential use cases in deployment. While this application direction is very promising empirically, foundation models are known to hallucinate and generate decisions that may sound reasonable, but are in fact poor. We argue there is a need to step back and simultaneously design systems that can quantify the certainty of a model's decision, and detect when it may be hallucinating. In this work, we discuss the current use cases of foundation models for decision-making tasks, provide a general definition for hallucinations with examples, discuss existing approaches to hallucination detection and mitigation with a focus on decision problems, and explore areas for further research in this exciting field.
comment: 31 pages, 2 tables
☆ Spatially temporally distributed informative path planning for multi-robot systems
This paper investigates the problem of informative path planning for a mobile robotic sensor network in spatially temporally distributed mapping. The robots are able to gather noisy measurements from an area of interest during their movements to build a Gaussian Process (GP) model of a spatio-temporal field. The model is then utilized to predict the spatio-temporal phenomenon at different points of interest. To spatially and temporally navigate the group of robots so that they can optimally acquire maximal information gains while their connectivity is preserved, we propose a novel multistep prediction informative path planning optimization strategy employing our newly defined local cost functions. By using the dual decomposition method, it is feasible and practical to effectively solve the optimization problem in a distributed manner. The proposed method was validated through synthetic experiments utilizing real-world data sets.
☆ Real-time Model Predictive Control with Zonotope-Based Neural Networks for Bipedal Social Navigation
This study addresses the challenge of bipedal navigation in a dynamic human-crowded environment, a research area that remains largely underexplored in the field of legged navigation. We propose two cascaded zonotope-based neural networks: a Pedestrian Prediction Network (PPN) for pedestrians' future trajectory prediction and an Ego-agent Social Network (ESN) for ego-agent social path planning. Representing future paths as zonotopes allows for efficient reachability-based planning and collision checking. The ESN is then integrated with a Model Predictive Controller (ESN-MPC) for footstep planning for our bipedal robot Digit designed by Agility Robotics. ESN-MPC solves for a collision-free optimal trajectory by optimizing through the gradients of ESN. ESN-MPC optimal trajectory is sent to the low-level controller for full-order simulation of Digit. The overall proposed framework is validated with extensive simulations on randomly generated initial settings with varying human crowd densities.
comment: 8 pages, 9 figures
☆ Towards Cooperative Maneuver Planning in Mixed Traffic at Urban Intersections
Connected automated driving promises a significant improvement of traffic efficiency and safety on highways and in urban areas. Apart from sharing of awareness and perception information over wireless communication links, cooperative maneuver planning may facilitate active guidance of connected automated vehicles at urban intersections. Research in automatic intersection management put forth a large body of works that mostly employ rule-based or optimization-based approaches primarily in fully automated simulated environments. In this work, we present two cooperative planning approaches that are capable of handling mixed traffic, i.e., the road being shared by automated vehicles and regular vehicles driven by humans. Firstly, we propose an optimization-based planner trained on real driving data that cyclically selects the most efficient out of multiple predicted coordinated maneuvers. Additionally, we present a cooperative planning approach based on graph-based reinforcement learning, which conquers the lack of ground truth data for cooperative maneuvers. We present evaluation results of both cooperative planners in high-fidelity simulation and real-world traffic. Simulative experiments in fully automated traffic and mixed traffic show that cooperative maneuver planning leads to less delay due to interaction and a reduced number of stops. In real-world experiments with three prototype connected automated vehicles in public traffic, both planners demonstrate their ability to perform efficient cooperative maneuvers.
comment: M. Klimke and M. Mertens are both first authors with equal contribution. 11 pages, 10 figures, 2 tables, submitted to IEEE Transactions on Intelligent Vehicles
☆ Producing and Leveraging Online Map Uncertainty in Trajectory Prediction CVPR 2024
High-definition (HD) maps have played an integral role in the development of modern autonomous vehicle (AV) stacks, albeit with high associated labeling and maintenance costs. As a result, many recent works have proposed methods for estimating HD maps online from sensor data, enabling AVs to operate outside of previously-mapped regions. However, current online map estimation approaches are developed in isolation of their downstream tasks, complicating their integration in AV stacks. In particular, they do not produce uncertainty or confidence estimates. In this work, we extend multiple state-of-the-art online map estimation methods to additionally estimate uncertainty and show how this enables more tightly integrating online mapping with trajectory forecasting. In doing so, we find that incorporating uncertainty yields up to 50% faster training convergence and up to 15% better prediction performance on the real-world nuScenes driving dataset.
comment: 14 pages, 14 figures, 6 tables. CVPR 2024
☆ AeroBridge: Autonomous Drone Handoff System for Emergency Battery Service
This paper proposes an Emergency Battery Service (EBS) for drones in which an EBS drone flies to a drone in the field with a depleted battery and transfers a fresh battery to the exhausted drone. The authors present a unique battery transfer mechanism and drone localization that uses the Cross Marker Position (CMP) method. The main challenges include a stable and balanced transfer that precisely localizes the receiver drone. The proposed EBS drone mitigates the effects of downwash due to the vertical proximity between the drones by implementing diagonal alignment with the receiver, reducing the distance to 0.5 m between the two drones. CFD analysis shows that diagonal instead of perpendicular alignment minimizes turbulence, and the authors verify the actual system for change in output airflow and thrust measurements. The CMP marker-based localization method enables position lock for the EBS drone with up to 0.9 cm accuracy. The performance of the transfer mechanism is validated experimentally by successful mid-air transfer in 5 seconds, where the EBS drone is within 0.5 m vertical distance from the receiver drone, wherein 4m/s turbulence does not affect the transfer process.
☆ Enhancing Visual Place Recognition via Fast and Slow Adaptive Biasing in Event Cameras
Event cameras are increasingly popular in robotics due to their beneficial features, such as low latency, energy efficiency, and high dynamic range. Nevertheless, their downstream task performance is greatly influenced by the optimization of bias parameters. These parameters, for instance, regulate the necessary change in light intensity to trigger an event, which in turn depends on factors such as the environment lighting and camera motion. This paper introduces feedback control algorithms that automatically tune the bias parameters through two interacting methods: 1) An immediate, on-the-fly fast adaptation of the refractory period, which sets the minimum interval between consecutive events, and 2) if the event rate exceeds the specified bounds even after changing the refractory period repeatedly, the controller adapts the pixel bandwidth and event thresholds, which stabilizes after a short period of noise events across all pixels (slow adaptation). Our evaluation focuses on the visual place recognition task, where incoming query images are compared to a given reference database. We conducted comprehensive evaluations of our algorithms' adaptive feedback control in real-time. To do so, we collected the QCR-Fast-and-Slow dataset that contains DAVIS346 event camera streams from 366 repeated traversals of a Scout Mini robot navigating through a 100 meter long indoor lab setting (totaling over 35km distance traveled) in varying brightness conditions with ground truth location information. Our proposed feedback controllers result in superior performance when compared to the standard bias settings and prior feedback control methods. Our findings also detail the impact of bias adjustments on task performance and feature ablation studies on the fast and slow adaptation mechanisms.
comment: 8 pages, 9 figures, paper under review
☆ Terrain-Attentive Learning for Efficient 6-DoF Kinodynamic Modeling on Vertically Challenging Terrain
Wheeled robots have recently demonstrated superior mechanical capability to traverse vertically challenging terrain (e.g., extremely rugged boulders comparable in size to the vehicles themselves). Negotiating such terrain introduces significant variations of vehicle pose in all six Degrees-of-Freedom (DoFs), leading to imbalanced contact forces, varying momentum, and chassis deformation due to non-rigid tires and suspensions. To autonomously navigate on vertically challenging terrain, all these factors need to be efficiently reasoned within limited onboard computation and strict real-time constraints. In this paper, we propose a 6-DoF kinodynamics learning approach that is attentive only to the specific underlying terrain critical to the current vehicle-terrain interaction, so that it can be efficiently queried in real-time motion planners onboard small robots. Physical experiment results show our Terrain-Attentive Learning demonstrates on average 51.1% reduction in model prediction error among all 6 DoFs compared to a state-of-the-art model for vertically challenging terrain.
☆ ASDF: Assembly State Detection Utilizing Late Fusion by Integrating 6D Pose Estimation
In medical and industrial domains, providing guidance for assembly processes is critical to ensure efficiency and safety. Errors in assembly can lead to significant consequences such as extended surgery times, and prolonged manufacturing or maintenance times in industry. Assembly scenarios can benefit from in-situ AR visualization to provide guidance, reduce assembly times and minimize errors. To enable in-situ visualization 6D pose estimation can be leveraged. Existing 6D pose estimation techniques primarily focus on individual objects and static captures. However, assembly scenarios have various dynamics including occlusion during assembly and dynamics in the assembly objects appearance. Existing work, combining object detection/6D pose estimation and assembly state detection focuses either on pure deep learning-based approaches, or limit the assembly state detection to building blocks. To address the challenges of 6D pose estimation in combination with assembly state detection, our approach ASDF builds upon the strengths of YOLOv8, a real-time capable object detection framework. We extend this framework, refine the object pose and fuse pose knowledge with network-detected pose information. Utilizing our late fusion in our Pose2State module results in refined 6D pose estimation and assembly state detection. By combining both pose and state information, our Pose2State module predicts the final assembly state with precision. Our evaluation on our ASDF dataset shows that our Pose2State module leads to an improved assembly state detection and that the improvement of the assembly state further leads to a more robust 6D pose estimation. Moreover, on the GBOT dataset, we outperform the pure deep learning-based network, and even outperform the hybrid and pure tracking-based approaches.
☆ SE(3) Linear Parameter Varying Dynamical Systems for Globally Asymptotically Stable End-Effector Control
Linear Parameter Varying Dynamical Systems (LPV-DS) encode trajectories into an autonomous first-order DS that enables reactive responses to perturbations, while ensuring globally asymptotic stability at the target. However, the current LPV-DS framework is established on Euclidean data only and has not been applicable to broader robotic applications requiring pose control. In this paper we present an extension to the current LPV-DS framework, named Quaternion-DS, which efficiently learns a DS-based motion policy for orientation. Leveraging techniques from differential geometry and Riemannian statistics, our approach properly handles the non-Euclidean orientation data in quaternion space, enabling the integration with positional control, namely SE(3) LPV-DS, so that the synergistic behaviour within the full SE(3) pose is preserved. Through simulation and real robot experiments, we validate our method, demonstrating its ability to efficiently and accurately reproduce the original SE(3) trajectory while exhibiting strong robustness to perturbations in task space.
☆ Bipedal Safe Navigation over Uncertain Rough Terrain: Unifying Terrain Mapping and Locomotion Stability
We study the problem of bipedal robot navigation in complex environments with uncertain and rough terrain. In particular, we consider a scenario in which the robot is expected to reach a desired goal location by traversing an environment with uncertain terrain elevation. Such terrain uncertainties induce not only untraversable regions but also robot motion perturbations. Thus, the problems of terrain mapping and locomotion stability are intertwined. We evaluate three different kernels for Gaussian process (GP) regression to learn the terrain elevation. We also learn the motion deviation resulting from both the terrain as well as the discrepancy between the reduced-order Prismatic Inverted Pendulum Model used for planning and the full-order locomotion dynamics. We propose a hierarchical locomotion-dynamics-aware sampling-based navigation planner. The global navigation planner plans a series of local waypoints to reach the desired goal locations while respecting locomotion stability constraints. Then, a local navigation planner is used to generate a sequence of dynamically feasible footsteps to reach local waypoints. We develop a novel trajectory evaluation metric to minimize motion deviation and maximize information gain of the terrain elevation map. We evaluate the efficacy of our planning framework on Digit bipedal robot simulation in MuJoCo.
comment: 9 pages, 7 figures
☆ Human Stress Response and Perceived Safety during Encounters with Quadruped Robots
Despite the rise of mobile robot deployments in home and work settings, perceived safety of users and bystanders is understudied in the human-robot interaction (HRI) literature. To address this, we present a study designed to identify elements of a human-robot encounter that correlate with observed stress response. Stress is a key component of perceived safety and is strongly associated with human physiological response. In this study a Boston Dynamics Spot and a Unitree Go1 navigate autonomously through a shared environment occupied by human participants wearing multimodal physiological sensors to track their electrocardiography (ECG) and electrodermal activity (EDA). The encounters are varied through several trials and participants self-rate their stress levels after each encounter. The study resulted in a multidimensional dataset archiving various objective and subjective aspects of a human-robot encounter, containing insights for understanding perceived safety in such encounters. To this end, acute stress responses were decoded from the human participants' ECG and EDA and compared across different human-robot encounter conditions. Statistical analysis of data indicate that on average (1) participants feel more stress during encounters compared to baselines, (2) participants feel more stress encountering multiple robots compared to a single robot and (3) participants stress increases during navigation behavior compared with search behavior.
comment: 7 pages, 7 figs, 5 tables
☆ Exploring CausalWorld: Enhancing robotic manipulation via knowledge transfer and curriculum learning
This study explores a learning-based tri-finger robotic arm manipulating task, which requires complex movements and coordination among the fingers. By employing reinforcement learning, we train an agent to acquire the necessary skills for proficient manipulation. To enhance the efficiency and effectiveness of the learning process, two knowledge transfer strategies, fine-tuning and curriculum learning, were utilized within the soft actor-critic architecture. Fine-tuning allows the agent to leverage pre-trained knowledge and adapt it to new tasks. Several variations like model transfer, policy transfer, and across-task transfer were implemented and evaluated. To eliminate the need for pretraining, curriculum learning decomposes the advanced task into simpler, progressive stages, mirroring how humans learn. The number of learning stages, the context of the sub-tasks, and the transition timing were found to be the critical design parameters. The key factors of two learning strategies and corresponding effects were explored in context-aware and context-unaware scenarios, enabling us to identify the scenarios where the methods demonstrate optimal performance, derive conclusive insights, and contribute to a broader range of learning-based engineering applications.
☆ Impact-Aware Bimanual Catching of Large-Momentum Objects
This paper investigates one of the most challenging tasks in dynamic manipulation -- catching large-momentum moving objects. Beyond the realm of quasi-static manipulation, dealing with highly dynamic objects can significantly improve the robot's capability of interacting with its surrounding environment. Yet, the inevitable motion mismatch between the fast moving object and the approaching robot will result in large impulsive forces, which lead to the unstable contacts and irreversible damage to both the object and the robot. To address the above problems, we propose an online optimization framework to: 1) estimate and predict the linear and angular motion of the object; 2) search and select the optimal contact locations across every surface of the object to mitigate impact through sequential quadratic programming (SQP); 3) simultaneously optimize the end-effector motion, stiffness, and contact force for both robots using multi-mode trajectory optimization (MMTO); and 4) realise the impact-aware catching motion on the compliant robotic system based on indirect force controller. We validate the impulse distribution, contact selection, and impact-aware MMTO algorithms in simulation and demonstrate the benefits of the proposed framework in real-world experiments including catching large-momentum moving objects with well-defined motion, constrained motion and free-flying motion.
☆ DASA: Delay-Adaptive Multi-Agent Stochastic Approximation
We consider a setting in which $N$ agents aim to speedup a common Stochastic Approximation (SA) problem by acting in parallel and communicating with a central server. We assume that the up-link transmissions to the server are subject to asynchronous and potentially unbounded time-varying delays. To mitigate the effect of delays and stragglers while reaping the benefits of distributed computation, we propose \texttt{DASA}, a Delay-Adaptive algorithm for multi-agent Stochastic Approximation. We provide a finite-time analysis of \texttt{DASA} assuming that the agents' stochastic observation processes are independent Markov chains. Significantly advancing existing results, \texttt{DASA} is the first algorithm whose convergence rate depends only on the mixing time $\tmix$ and on the average delay $\tau_{avg}$ while jointly achieving an $N$-fold convergence speedup under Markovian sampling. Our work is relevant for various SA applications, including multi-agent and distributed temporal difference (TD) learning, Q-learning and stochastic optimization with correlated data.
☆ TwoStep: Multi-agent Task Planning using Classical Planners and Large Language Models
Classical planning formulations like the Planning Domain Definition Language (PDDL) admit action sequences guaranteed to achieve a goal state given an initial state if any are possible. However, reasoning problems defined in PDDL do not capture temporal aspects of action taking, for example that two agents in the domain can execute an action simultaneously if postconditions of each do not interfere with preconditions of the other. A human expert can decompose a goal into largely independent constituent parts and assign each agent to one of these subgoals to take advantage of simultaneous actions for faster execution of plan steps, each using only single agent planning. By contrast, large language models (LLMs) used for directly inferring plan steps do not guarantee execution success, but do leverage commonsense reasoning to assemble action sequences. We combine the strengths of classical planning and LLMs by approximating human intuitions for two-agent planning goal decomposition. We demonstrate that LLM-based goal decomposition leads to faster planning times than solving multi-agent PDDL problems directly while simultaneously achieving fewer plan execution steps than a single agent plan alone and preserving execution success. Additionally, we find that LLM-based approximations of subgoals can achieve similar multi-agent execution steps than those specified by human experts. Website and resources at https://glamor-usc.github.io/twostep
comment: 12 pages
☆ Temporal and Semantic Evaluation Metrics for Foundation Models in Post-Hoc Analysis of Robotic Sub-tasks IROS 2024
Recent works in Task and Motion Planning (TAMP) show that training control policies on language-supervised robot trajectories with quality labeled data markedly improves agent task success rates. However, the scarcity of such data presents a significant hurdle to extending these methods to general use cases. To address this concern, we present an automated framework to decompose trajectory data into temporally bounded and natural language-based descriptive sub-tasks by leveraging recent prompting strategies for Foundation Models (FMs) including both Large Language Models (LLMs) and Vision Language Models (VLMs). Our framework provides both time-based and language-based descriptions for lower-level sub-tasks that comprise full trajectories. To rigorously evaluate the quality of our automatic labeling framework, we contribute an algorithm SIMILARITY to produce two novel metrics, temporal similarity and semantic similarity. The metrics measure the temporal alignment and semantic fidelity of language descriptions between two sub-task decompositions, namely an FM sub-task decomposition prediction and a ground-truth sub-task decomposition. We present scores for temporal similarity and semantic similarity above 90%, compared to 30% of a randomized baseline, for multiple robotic environments, demonstrating the effectiveness of our proposed framework. Our results enable building diverse, large-scale, language-supervised datasets for improved robotic TAMP.
comment: 8 pages, 3 figures. IROS 2024 Submission
☆ Speeding Up Path Planning via Reinforcement Learning in MCTS for Automated Parking
In this paper, we address a method that integrates reinforcement learning into the Monte Carlo tree search to boost online path planning under fully observable environments for automated parking tasks. Sampling-based planning methods under high-dimensional space can be computationally expensive and time-consuming. State evaluation methods are useful by leveraging the prior knowledge into the search steps, making the process faster in a real-time system. Given the fact that automated parking tasks are often executed under complex environments, a solid but lightweight heuristic guidance is challenging to compose in a traditional analytical way. To overcome this limitation, we propose a reinforcement learning pipeline with a Monte Carlo tree search under the path planning framework. By iteratively learning the value of a state and the best action among samples from its previous cycle's outcomes, we are able to model a value estimator and a policy generator for given states. By doing that, we build up a balancing mechanism between exploration and exploitation, speeding up the path planning process while maintaining its quality without using human expert driver data.
☆ PROSPECT: Precision Robot Spectroscopy Exploration and Characterization Tool
Near Infrared (NIR) spectroscopy is widely used in industrial quality control and automation to test the purity and material quality of items. In this research, we propose a novel sensorized end effector and acquisition strategy to capture spectral signatures from objects and register them with a 3D point cloud. Our methodology first takes a 3D scan of an object generated by a time-of-flight depth camera and decomposes the object into a series of planned viewpoints covering the surface. We generate motion plans for a robot manipulator and end-effector to visit these viewpoints while maintaining a fixed distance and surface normal to ensure maximal spectral signal quality enabled by the spherical motion of the end-effector. By continuously acquiring surface reflectance values as the end-effector scans the target object, the autonomous system develops a four-dimensional model of the target object: position in an R^3 coordinate frame, and a wavelength vector denoting the associated spectral signature. We demonstrate this system in building spectral-spatial object profiles of increasingly complex geometries. As a point of comparison, we show our proposed system and spectral acquisition planning yields more consistent signal signals than naive point scanning strategies for capturing spectral information over complex surface geometries. Our work represents a significant step towards high-resolution spectral-spatial sensor fusion for automated quality assessment.
☆ Dyna-LfLH: Learning Agile Navigation in Dynamic Environments from Learned Hallucination IROS
This paper presents a self-supervised learning method to safely learn a motion planner for ground robots to navigate environments with dense and dynamic obstacles. When facing highly-cluttered, fast-moving, hard-to-predict obstacles, classical motion planners may not be able to keep up with limited onboard computation. For learning-based planners, high-quality demonstrations are difficult to acquire for imitation learning while reinforcement learning becomes inefficient due to the high probability of collision during exploration. To safely and efficiently provide training data, the Learning from Hallucination (LfH) approaches synthesize difficult navigation environments based on past successful navigation experiences in relatively easy or completely open ones, but unfortunately cannot address dynamic obstacles. In our new Dynamic Learning from Learned Hallucination (Dyna-LfLH), we design and learn a novel latent distribution and sample dynamic obstacles from it, so the generated training data can be used to learn a motion planner to navigate in dynamic environments. Dyna-LfLH is evaluated on a ground robot in both simulated and physical environments and achieves up to 25% better success rate compared to baselines.
comment: Submitted to International Conference on Intelligent Robots and Systems (IROS) 2024
☆ Multi-Contact Inertial Estimation and Localization in Legged Robots
Optimal estimation is a promising tool for multi-contact inertial estimation and localization. To harness its advantages in robotics, it is crucial to solve these large and challenging optimization problems efficiently. To tackle this, we (i) develop a multiple-shooting solver that exploits both temporal and parametric structures through a parametrized Riccati recursion. Additionally, we (ii) propose an inertial local manifold that ensures its full physical consistency. It also enhances convergence compared to the singularity-free log-Cholesky approach. To handle its singularities, we (iii) introduce a nullspace approach in our optimal estimation solver. We (iv) finally develop the analytical derivatives of contact dynamics for both inertial parametrizations. Our framework can successfully solve estimation problems for complex maneuvers such as brachiation in humanoids. We demonstrate its numerical capabilities across various robotics tasks and its benefits in experimental trials with the Go1 robot.
☆ Hearing the shape of an arena with spectral swarm robotics
Swarm robotics promises adaptability to unknown situations and robustness against failures. However, it still struggles with global tasks that require understanding the broader context in which the robots operate, such as identifying the shape of the arena in which the robots are embedded. Biological swarms, such as shoals of fish, flocks of birds, and colonies of insects, routinely solve global geometrical problems through the diffusion of local cues. This paradigm can be explicitly described by mathematical models that could be directly computed and exploited by a robotic swarm. Diffusion over a domain is mathematically encapsulated by the Laplacian, a linear operator that measures the local curvature of a function. Crucially the geometry of a domain can generally be reconstructed from the eigenspectrum of its Laplacian. Here we introduce spectral swarm robotics where robots diffuse information to their neighbors to emulate the Laplacian operator - enabling them to "hear" the spectrum of their arena. We reveal a universal scaling that links the optimal number of robots (a global parameter) with their optimal radius of interaction (a local parameter). We validate experimentally spectral swarm robotics under challenging conditions with the one-shot classification of arena shapes using a sparse swarm of Kilobots. Spectral methods can assist with challenging tasks where robots need to build an emergent consensus on their environment, such as adaptation to unknown terrains, division of labor, or quorum sensing. Spectral methods may extend beyond robotics to analyze and coordinate swarms of agents of various natures, such as traffic or crowds, and to better understand the long-range dynamics of natural systems emerging from short-range interactions.
☆ Adaptive Step Duration for Precise Foot Placement: Achieving Robust Bipedal Locomotion on Terrains with Restricted Footholds
This paper introduces a novel multi-step preview foot placement planning algorithm designed to enhance the robustness of bipedal robotic walking across challenging terrains with restricted footholds. Traditional one-step preview planning struggles to maintain stability when stepping areas are severely limited, such as with random stepping stones. In this work, we developed a discrete-time Model Predictive Control (MPC) based on the step-to-step discrete evolution of the Divergent Component of Motion (DCM) of bipedal locomotion. This approach adaptively changes the step duration for optimal foot placement under constraints, thereby ensuring the robot's operational viability over multiple future steps and significantly improving its ability to navigate through environments with tight constraints on possible footholds. The effectiveness of this planning algorithm is demonstrated through simulations that include a variety of complex stepping-stone configurations and external perturbations. These tests underscore the algorithm's improved performance for navigating foothold-restricted environments, even with the presence of external disturbances.
comment: 8 pages, 8 figures, submitted to CDC 2024, for associated simulation video, see https://youtu.be/2jhikPlZmbE
☆ Grounding Language Plans in Demonstrations Through Counterfactual Perturbations
Grounding the common-sense reasoning of Large Language Models in physical domains remains a pivotal yet unsolved problem for embodied AI. Whereas prior works have focused on leveraging LLMs directly for planning in symbolic spaces, this work uses LLMs to guide the search of task structures and constraints implicit in multi-step demonstrations. Specifically, we borrow from manipulation planning literature the concept of mode families, which group robot configurations by specific motion constraints, to serve as an abstraction layer between the high-level language representations of an LLM and the low-level physical trajectories of a robot. By replaying a few human demonstrations with synthetic perturbations, we generate coverage over the demonstrations' state space with additional successful executions as well as counterfactuals that fail the task. Our explanation-based learning framework trains an end-to-end differentiable neural network to predict successful trajectories from failures and as a by-product learns classifiers that ground low-level states and images in mode families without dense labeling. The learned grounding classifiers can further be used to translate language plans into reactive policies in the physical domain in an interpretable manner. We show our approach improves the interpretability and reactivity of imitation learning through 2D navigation and simulated and real robot manipulation tasks. Website: https://sites.google.com/view/grounding-plans
☆ Vision-Based Dexterous Motion Planning by Dynamic Movement Primitives with Human Hand Demonstration
This paper proposes a vision-based framework for a 7-degree-of-freedom robotic manipulator, with the primary objective of facilitating its capacity to acquire information from human hand demonstrations for the execution of dexterous pick-and-place tasks. Most existing works only focus on the position demonstration without considering the orientations. In this paper, by employing a single depth camera, MediaPipe is applied to generate the three-dimensional coordinates of a human hand, thereby comprehensively recording the hand's motion, encompassing the trajectory of the wrist, orientation of the hand, and the grasp motion. A mean filter is applied during data pre-processing to smooth the raw data. The demonstration is designed to pick up an object at a specific angle, navigate around obstacles in its path and subsequently, deposit it within a sloped container. The robotic system demonstrates its learning capabilities, facilitated by the implementation of Dynamic Movement Primitives, enabling the assimilation of user actions into its trajectories with different start and end poi
☆ Berry Twist: a Twisting-Tube Soft Robotic Gripper for Blackberry Harvesting
As global demand for fruits and vegetables continues to rise, the agricultural industry faces challenges in securing adequate labor. Robotic harvesting devices offer a promising solution to solve this issue. However, harvesting delicate fruits, notably blackberries, poses unique challenges due to their fragility. This study introduces and evaluates a prototype robotic gripper specifically designed for blackberry harvesting. The gripper features an innovative fabric tube mechanism employing motorized twisting action to gently envelop the fruit, ensuring uniform pressure application and minimizing damage. Three types of tubes were developed, varying in elasticity and compressibility using foam padding, spandex, and food-safe cotton cheesecloth. Performance testing focused on assessing each gripper's ability to detach and release blackberries, with emphasis on quantifying damage rates. Results indicate the proposed gripper achieved an 82% success rate in detaching blackberries and a 95% success rate in releasing them, showcasing the promised potential for robotic harvesting applications.
☆ A Comparative Analysis of Visual Odometry in Virtual and Real-World Railways Environments
Perception tasks play a crucial role in the development of automated operations and systems across multiple application fields. In the railway transportation domain, these tasks can improve the safety, reliability, and efficiency of various perations, including train localization, signal recognition, and track discrimination. However, collecting considerable and precisely labeled datasets for testing such novel algorithms poses extreme challenges in the railway environment due to the severe restrictions in accessing the infrastructures and the practical difficulties associated with properly equipping trains with the required sensors, such as cameras and LiDARs. The remarkable innovations of graphic engine tools offer new solutions to craft realistic synthetic datasets. To illustrate the advantages of employing graphic simulation for early-stage testing of perception tasks in the railway domain, this paper presents a comparative analysis of the performance of a SLAM algorithm applied both in a virtual synthetic environment and a real-world scenario. The analysis leverages virtual railway environments created with the latest version of Unreal Engine, facilitating data collection and allowing the examination of challenging scenarios, including low-visibility, dangerous operational modes, and complex environments. The results highlight the feasibility and potentiality of graphic simulation to advance perception tasks in the railway domain.
Trajectory Optimization with Global Yaw Parameterization for Field-of-View Constrained Autonomous Flight
Trajectory generation for quadrotors with limited field-of-view sensors has numerous applications such as aerial exploration, coverage, inspection, videography, and target tracking. Most previous works simplify the task of optimizing yaw trajectories by either aligning the heading of the robot with its velocity, or potentially restricting the feasible space of candidate trajectories by using a limited yaw domain to circumvent angular singularities. In this paper, we propose a novel \textit{global} yaw parameterization method for trajectory optimization that allows a 360-degree yaw variation as demanded by the underlying algorithm. This approach effectively bypasses inherent singularities by including supplementary quadratic constraints and transforming the final decision variables into the desired state representation. This method significantly reduces the needed control effort, and improves optimization feasibility. Furthermore, we apply the method to several examples of different applications that require jointly optimizing over both the yaw and position trajectories. Ultimately, we present a comprehensive numerical analysis and evaluation of our proposed method in both simulation and real-world experiments.
Calib3D: Calibrating Model Preferences for Reliable 3D Scene Understanding
Safety-critical 3D scene understanding tasks necessitate not only accurate but also confident predictions from 3D perception models. This study introduces Calib3D, a pioneering effort to benchmark and scrutinize the reliability of 3D scene understanding models from an uncertainty estimation viewpoint. We comprehensively evaluate 28 state-of-the-art models across 10 diverse 3D datasets, uncovering insightful phenomena that cope with both the aleatoric and epistemic uncertainties in 3D scene understanding. We discover that despite achieving impressive levels of accuracy, existing models frequently fail to provide reliable uncertainty estimates -- a pitfall that critically undermines their applicability in safety-sensitive contexts. Through extensive analysis of key factors such as network capacity, LiDAR representations, rasterization resolutions, and 3D data augmentation techniques, we correlate these aspects directly with the model calibration efficacy. Furthermore, we introduce DeptS, a novel depth-aware scaling approach aimed at enhancing 3D model calibration. Extensive experiments across a wide range of configurations validate the superiority of our method. We hope this work could serve as a cornerstone for fostering reliable 3D scene understanding. Code and benchmark toolkits are publicly available.
comment: Preprint; 37 pages, 8 figures, 11 tables; Code at https://github.com/ldkong1205/Calib3D
☆ Optimizing LiDAR Placements for Robust Driving Perception in Adverse Conditions
The robustness of driving perception systems under unprecedented conditions is crucial for safety-critical usages. Latest advancements have prompted increasing interests towards multi-LiDAR perception. However, prevailing driving datasets predominantly utilize single-LiDAR systems and collect data devoid of adverse conditions, failing to capture the complexities of real-world environments accurately. Addressing these gaps, we proposed Place3D, a full-cycle pipeline that encompasses LiDAR placement optimization, data generation, and downstream evaluations. Our framework makes three appealing contributions. 1) To identify the most effective configurations for multi-LiDAR systems, we introduce a Surrogate Metric of the Semantic Occupancy Grids (M-SOG) to evaluate LiDAR placement quality. 2) Leveraging the M-SOG metric, we propose a novel optimization strategy to refine multi-LiDAR placements. 3) Centered around the theme of multi-condition multi-LiDAR perception, we collect a 364,000-frame dataset from both clean and adverse conditions. Extensive experiments demonstrate that LiDAR placements optimized using our approach outperform various baselines. We showcase exceptional robustness in both 3D object detection and LiDAR semantic segmentation tasks, under diverse adverse weather and sensor failure conditions. Code and benchmark toolkit are publicly available.
comment: Preprint; 40 pages, 11 figures, 15 tables; Code at https://github.com/ywyeli/Place3D
☆ DriveCoT: Integrating Chain-of-Thought Reasoning with End-to-End Driving
End-to-end driving has made significant progress in recent years, demonstrating benefits such as system simplicity and competitive driving performance under both open-loop and closed-loop settings. Nevertheless, the lack of interpretability and controllability in its driving decisions hinders real-world deployment for end-to-end driving systems. In this paper, we collect a comprehensive end-to-end driving dataset named DriveCoT, leveraging the CARLA simulator. It contains sensor data, control decisions, and chain-of-thought labels to indicate the reasoning process. We utilize the challenging driving scenarios from the CARLA leaderboard 2.0, which involve high-speed driving and lane-changing, and propose a rule-based expert policy to control the vehicle and generate ground truth labels for its reasoning process across different driving aspects and the final decisions. This dataset can serve as an open-loop end-to-end driving benchmark, enabling the evaluation of accuracy in various chain-of-thought aspects and the final decision. In addition, we propose a baseline model called DriveCoT-Agent, trained on our dataset, to generate chain-of-thought predictions and final decisions. The trained model exhibits strong performance in both open-loop and closed-loop evaluations, demonstrating the effectiveness of our proposed dataset.
Visual Whole-Body Control for Legged Loco-Manipulation
We study the problem of mobile manipulation using legged robots equipped with an arm, namely legged loco-manipulation. The robot legs, while usually utilized for mobility, offer an opportunity to amplify the manipulation capabilities by conducting whole-body control. That is, the robot can control the legs and the arm at the same time to extend its workspace. We propose a framework that can conduct the whole-body control autonomously with visual observations. Our approach, namely \ourFull~(\our), is composed of a low-level policy using all degrees of freedom to track the end-effector manipulator position and a high-level policy proposing the end-effector position based on visual inputs. We train both levels of policies in simulation and perform Sim2Real transfer for real robot deployment. We perform extensive experiments and show significant improvements over baselines in picking up diverse objects in different configurations (heights, locations, orientations) and environments. Project page: https://wholebody-b1.github.io
comment: The first two authors contribute equally. Project page: https://wholebody-b1.github.io
☆ Bayesian Methods for Trust in Collaborative Multi-Agent Autonomy
Multi-agent, collaborative sensor fusion is a vital component of a multi-national intelligence toolkit. In safety-critical and/or contested environments, adversaries may infiltrate and compromise a number of agents. We analyze state of the art multi-target tracking algorithms under this compromised agent threat model. We prove that the track existence probability test ("track score") is significantly vulnerable to even small numbers of adversaries. To add security awareness, we design a trust estimation framework using hierarchical Bayesian updating. Our framework builds beliefs of trust on tracks and agents by mapping sensor measurements to trust pseudomeasurements (PSMs) and incorporating prior trust beliefs in a Bayesian context. In case studies, our trust estimation algorithm accurately estimates the trustworthiness of tracks/agents, subject to observability limitations.
☆ Learning Symbolic and Subsymbolic Temporal Task Constraints from Bimanual Human Demonstrations IROS 2024
Learning task models of bimanual manipulation from human demonstration and their execution on a robot should take temporal constraints between actions into account. This includes constraints on (i) the symbolic level such as precedence relations or temporal overlap in the execution, and (ii) the subsymbolic level such as the duration of different actions, or their starting and end points in time. Such temporal constraints are crucial for temporal planning, reasoning, and the exact timing for the execution of bimanual actions on a bimanual robot. In our previous work, we addressed the learning of temporal task constraints on the symbolic level and demonstrated how a robot can leverage this knowledge to respond to failures during execution. In this work, we propose a novel model-driven approach for the combined learning of symbolic and subsymbolic temporal task constraints from multiple bimanual human demonstrations. Our main contributions are a subsymbolic foundation of a temporal task model that describes temporal nexuses of actions in the task based on distributions of temporal differences between semantic action keypoints, as well as a method based on fuzzy logic to derive symbolic temporal task constraints from this representation. This complements our previous work on learning comprehensive temporal task models by integrating symbolic and subsymbolic information based on a subsymbolic foundation, while still maintaining the symbolic expressiveness of our previous approach. We compare our proposed approach with our previous pure-symbolic approach and show that we can reproduce and even outperform it. Additionally, we show how the subsymbolic temporal task constraints can synchronize otherwise unimanual movement primitives for bimanual behavior on a humanoid robot.
comment: 8 pages, submitted to IROS 2024
☆ DHP-Mapping: A Dense Panoptic Mapping System with Hierarchical World Representation and Label Optimization Techniques IROS 2024
Maps provide robots with crucial environmental knowledge, thereby enabling them to perform interactive tasks effectively. Easily accessing accurate abstract-to-detailed geometric and semantic concepts from maps is crucial for robots to make informed and efficient decisions. To comprehensively model the environment and effectively manage the map data structure, we propose DHP-Mapping, a dense mapping system that utilizes multiple Truncated Signed Distance Field (TSDF) submaps and panoptic labels to hierarchically model the environment. The output map is able to maintain both voxel- and submap-level metric and semantic information. Two modules are presented to enhance the mapping efficiency and label consistency: (1) an inter-submaps label fusion strategy to eliminate duplicate points across submaps and (2) a conditional random field (CRF) based approach to enhance panoptic labels through object label comprehension and contextual information. We conducted experiments with two public datasets including indoor and outdoor scenarios. Our system performs comparably to state-of-the-art (SOTA) methods across geometry and label accuracy evaluation metrics. The experiment results highlight the effectiveness and scalability of our system, as it is capable of constructing precise geometry and maintaining consistent panoptic labels. Our code is publicly available at https://github.com/hutslib/DHP-Mapping.
comment: Submit to IROS 2024. Project website https://github.com/hutslib/DHP-Mapping
☆ Proprioception Is All You Need: Terrain Classification for Boreal Forests IROS 2024
Recent works in field robotics highlighted the importance of resiliency against different types of terrains. Boreal forests, in particular, are home to many mobility-impeding terrains that should be considered for off-road autonomous navigation. Also, being one of the largest land biomes on Earth, boreal forests are an area where autonomous vehicles are expected to become increasingly common. In this paper, we address this issue by introducing BorealTC, a publicly available dataset for proprioceptive-based terrain classification (TC). Recorded with a Husky A200, our dataset contains 116 min of Inertial Measurement Unit (IMU), motor current, and wheel odometry data, focusing on typical boreal forest terrains, notably snow, ice, and silty loam. Combining our dataset with another dataset from the state-of-the-art, we evaluate both a Convolutional Neural Network (CNN) and the novel state space model (SSM)-based Mamba architecture on a TC task. Interestingly, we show that while CNN outperforms Mamba on each separate dataset, Mamba achieves greater accuracy when trained on a combination of both. In addition, we demonstrate that Mamba's learning capacity is greater than a CNN for increasing amounts of data. We show that the combination of two TC datasets yields a latent space that can be interpreted with the properties of the terrains. We also discuss the implications of merging datasets on classification. Our source code and dataset are publicly available online: https://github.com/norlab-ulaval/BorealTC.
comment: Submitted to the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)
☆ TAIL: A Terrain-Aware Multi-Modal SLAM Dataset for Robot Locomotion in Deformable Granular Environments IEEE Robotics and Automation Letters
Terrain-aware perception holds the potential to improve the robustness and accuracy of autonomous robot navigation in the wilds, thereby facilitating effective off-road traversals. However, the lack of multi-modal perception across various motion patterns hinders the solutions of Simultaneous Localization And Mapping (SLAM), especially when confronting non-geometric hazards in demanding landscapes. In this paper, we first propose a Terrain-Aware multI-modaL (TAIL) dataset tailored to deformable and sandy terrains. It incorporates various types of robotic proprioception and distinct ground interactions for the unique challenges and benchmark of multi-sensor fusion SLAM. The versatile sensor suite comprises stereo frame cameras, multiple ground-pointing RGB-D cameras, a rotating 3D LiDAR, an IMU, and an RTK device. This ensemble is hardware-synchronized, well-calibrated, and self-contained. Utilizing both wheeled and quadrupedal locomotion, we efficiently collect comprehensive sequences to capture rich unstructured scenarios. It spans the spectrum of scope, terrain interactions, scene changes, ground-level properties, and dynamic robot characteristics. We benchmark several state-of-the-art SLAM methods against ground truth and provide performance validations. Corresponding challenges and limitations are also reported. All associated resources are accessible upon request at \url{https://tailrobot.github.io/}.
comment: Submitted to IEEE Robotics and Automation Letters
☆ A Semi-Lagrangian Approach for Time and Energy Path Planning Optimization in Static Flow Fields
Efficient path planning for autonomous mobile robots is a critical problem across numerous domains, where optimizing both time and energy consumption is paramount. This paper introduces a novel methodology that considers the dynamic influence of an environmental flow field and considers geometric constraints, including obstacles and forbidden zones, enriching the complexity of the planning problem. We formulate it as a multi-objective optimal control problem, propose a novel transformation called Harmonic Transformation, and apply a semi-Lagrangian scheme to solve it. The set of Pareto efficient solutions is obtained considering two distinct approaches: a deterministic method and an evolutionary-based one, both of which are designed to make use of the proposed Harmonic Transformation. Through an extensive analysis of these approaches, we demonstrate their efficacy in finding optimized paths.
comment: 12 pages, initial paper submission; Preprint submitted to the IEEE Transactions on Intelligent Transportation Systems
☆ Exploiting Priors from 3D Diffusion Models for RGB-Based One-Shot View Planning IROS 2024
Object reconstruction is relevant for many autonomous robotic tasks that require interaction with the environment. A key challenge in such scenarios is planning view configurations to collect informative measurements for reconstructing an initially unknown object. One-shot view planning enables efficient data collection by predicting view configurations and planning the globally shortest path connecting all views at once. However, geometric priors about the object are required to conduct one-shot view planning. In this work, we propose a novel one-shot view planning approach that utilizes the powerful 3D generation capabilities of diffusion models as priors. By incorporating such geometric priors into our pipeline, we achieve effective one-shot view planning starting with only a single RGB image of the object to be reconstructed. Our planning experiments in simulation and real-world setups indicate that our approach balances well between object reconstruction quality and movement cost.
comment: Sicong Pan and Liren Jin have equal contribution. Submitted to IROS 2024
☆ CurbNet: Curb Detection Framework Based on LiDAR Point Cloud Segmentation
Curb detection is an important function in intelligent driving and can be used to determine drivable areas of the road. However, curbs are difficult to detect due to the complex road environment. This paper introduces CurbNet, a novel framework for curb detection, leveraging point cloud segmentation. Addressing the dearth of comprehensive curb datasets and the absence of 3D annotations, we have developed the 3D-Curb dataset, encompassing 7,100 frames, which represents the largest and most categorically diverse collection of curb point clouds currently available. Recognizing that curbs are primarily characterized by height variations, our approach harnesses spatially-rich 3D point clouds for training. To tackle the challenges presented by the uneven distribution of curb features on the xy-plane and their reliance on z-axis high-frequency features, we introduce the multi-scale and channel attention (MSCA) module, a bespoke solution designed to optimize detection performance. Moreover, we propose an adaptive weighted loss function group, specifically formulated to counteract the imbalance in the distribution of curb point clouds relative to other categories. Our extensive experimentation on 2 major datasets has yielded results that surpass existing benchmarks set by leading curb detection and point cloud segmentation models. By integrating multi-clustering and curve fitting techniques in our post-processing stage, we have substantially reduced noise in curb detection, thereby enhancing precision to 0.8744. Notably, CurbNet has achieved an exceptional average metrics of over 0.95 at a tolerance of just 0.15m, thereby establishing a new benchmark. Furthermore, corroborative real-world experiments and dataset analyzes mutually validate each other, solidifying CurbNet's superior detection proficiency and its robust generalizability.
☆ DBPF: A Framework for Efficient and Robust Dynamic Bin-Picking
Efficiency and reliability are critical in robotic bin-picking as they directly impact the productivity of automated industrial processes. However, traditional approaches, demanding static objects and fixed collisions, lead to deployment limitations, operational inefficiencies, and process unreliability. This paper introduces a Dynamic Bin-Picking Framework (DBPF) that challenges traditional static assumptions. The DBPF endows the robot with the reactivity to pick multiple moving arbitrary objects while avoiding dynamic obstacles, such as the moving bin. Combined with scene-level pose generation, the proposed pose selection metric leverages the Tendency-Aware Manipulability Network optimizing suction pose determination. Heuristic task-specific designs like velocity-matching, dynamic obstacle avoidance, and the resight policy, enhance the picking success rate and reliability. Empirical experiments demonstrate the importance of these components. Our method achieves an average 84% success rate, surpassing the 60% of the most comparable baseline, crucially, with zero collisions. Further evaluations under diverse dynamic scenarios showcase DBPF's robust performance in dynamic bin-picking. Results suggest that our framework offers a promising solution for efficient and reliable robotic bin-picking under dynamics.
comment: 8 pages, 5 figures. This paper has been accepted by IEEE RA-L on 2024-03-24. See the supplementary video at youtube: https://youtu.be/n5af2VsKhkg
☆ Visual Action Planning with Multiple Heterogeneous Agents
Visual planning methods are promising to handle complex settings where extracting the system state is challenging. However, none of the existing works tackles the case of multiple heterogeneous agents which are characterized by different capabilities and/or embodiment. In this work, we propose a method to realize visual action planning in multi-agent settings by exploiting a roadmap built in a low-dimensional structured latent space and used for planning. To enable multi-agent settings, we infer possible parallel actions from a dataset composed of tuples associated with individual actions. Next, we evaluate feasibility and cost of them based on the capabilities of the multi-agent system and endow the roadmap with this information, building a capability latent space roadmap (C-LSR). Additionally, a capability suggestion strategy is designed to inform the human operator about possible missing capabilities when no paths are found. The approach is validated in a simulated burger cooking task and a real-world box packing task.
☆ Low-Cost Teleoperation with Haptic Feedback through Vision-based Tactile Sensors for Rigid and Soft Object Manipulation
Haptic feedback is essential for humans to successfully perform complex and delicate manipulation tasks. A recent rise in tactile sensors has enabled robots to leverage the sense of touch and expand their capability drastically. However, many tasks still need human intervention/guidance. For this reason, we present a teleoperation framework designed to provide haptic feedback to human operators based on the data from camera-based tactile sensors mounted on the robot gripper. Partial autonomy is introduced to prevent slippage of grasped objects during task execution. Notably, we rely exclusively on low-cost off-the-shelf hardware to realize an affordable solution. We demonstrate the versatility of the framework on nine different objects ranging from rigid to soft and fragile ones, using three different operators on real hardware.
comment: https://vision-tactile-manip.github.io/teleop/
☆ ProIn: Learning to Predict Trajectory Based on Progressive Interactions for Autonomous Driving
Accurate motion prediction of pedestrians, cyclists, and other surrounding vehicles (all called agents) is very important for autonomous driving. Most existing works capture map information through an one-stage interaction with map by vector-based attention, to provide map constraints for social interaction and multi-modal differentiation. However, these methods have to encode all required map rules into the focal agent's feature, so as to retain all possible intentions' paths while at the meantime to adapt to potential social interaction. In this work, a progressive interaction network is proposed to enable the agent's feature to progressively focus on relevant maps, in order to better learn agents' feature representation capturing the relevant map constraints. The network progressively encode the complex influence of map constraints into the agent's feature through graph convolutions at the following three stages: after historical trajectory encoder, after social interaction, and after multi-modal differentiation. In addition, a weight allocation mechanism is proposed for multi-modal training, so that each mode can obtain learning opportunities from a single-mode ground truth. Experiments have validated the superiority of progressive interactions to the existing one-stage interaction, and demonstrate the effectiveness of each component. Encouraging results were obtained in the challenging benchmarks.
♻ ☆ A Modular Pneumatic Soft Gripper Design for Aerial Grasping and Landing
Aerial robots have garnered significant attention due to their potential applications in various industries, such as inspection, search and rescue, and drone delivery. Successful missions often depend on the ability of these robots to grasp and land effectively. This paper presents a novel modular soft gripper design tailored explicitly for aerial grasping and landing operations. The proposed modular pneumatic soft gripper incorporates a feed-forward proportional controller to regulate pressure, enabling compliant gripping capabilities. The modular connectors of the soft fingers offer two configurations for the 4-tip soft gripper, H-base (cylindrical) and X-base (spherical), allowing adaptability to different target objects. Additionally, the gripper can serve as a soft landing gear when deflated, eliminating the need for an extra landing gear. This design reduces weight, simplifies aerial manipulation control, and enhances flight efficiency. We demonstrate the efficacy of indoor aerial grasping and achieve a maximum payload of 217 g using the proposed soft aerial vehicle and its H-base pneumatic soft gripper (808 g).
comment: 7 pages, 13 figures, accepted by IEEE RoboSoft 2024
♻ ☆ Robust Integral Consensus Control of Multi-Agent Networks Perturbed by Matched and Unmatched Disturbances: The Case of Directed Graphs
This work presents a new method to design consensus controllers for perturbed double integrator systems whose interconnection is described by a directed graph containing a rooted spanning tree. We propose new robust controllers to solve the consensus and synchronization problems when the systems are under the effects of matched and unmatched disturbances. In both problems, we present simple continuous controllers, whose integral actions allow us to handle the disturbances. A rigorous stability analysis based on Lyapunov's direct method for unperturbed networked systems is presented. To assess the performance of our result, a representative simulation study is presented.
♻ ☆ A Group Theoretic Metric for Robot State Estimation Leveraging Chebyshev Interpolation ICRA 2024
We propose a new metric for robot state estimation based on the recently introduced $\text{SE}_2(3)$ Lie group definition. Our metric is related to prior metrics for SLAM but explicitly takes into account the linear velocity of the state estimate, improving over current pose-based trajectory analysis. This has the benefit of providing a single, quantitative metric to evaluate state estimation algorithms against, while being compatible with existing tools and libraries. Since ground truth data generally consists of pose data from motion capture systems, we also propose an approach to compute the ground truth linear velocity based on polynomial interpolation. Using Chebyshev interpolation and a pseudospectral parameterization, we can accurately estimate the ground truth linear velocity of the trajectory in an optimal fashion with best approximation error. We demonstrate how this approach performs on multiple robotic platforms where accurate state estimation is vital, and compare it to alternative approaches such as finite differences. The pseudospectral parameterization also provides a means of trajectory data compression as an additional benefit. Experimental results show our method provides a valid and accurate means of comparing state estimation systems, which is also easy to interpret and report.
comment: Accepted to ICRA 2024
♻ ☆ I-PHYRE: Interactive Physical Reasoning ICLR 2024
Current evaluation protocols predominantly assess physical reasoning in stationary scenes, creating a gap in evaluating agents' abilities to interact with dynamic events. While contemporary methods allow agents to modify initial scene configurations and observe consequences, they lack the capability to interact with events in real time. To address this, we introduce I-PHYRE, a framework that challenges agents to simultaneously exhibit intuitive physical reasoning, multi-step planning, and in-situ intervention. Here, intuitive physical reasoning refers to a quick, approximate understanding of physics to address complex problems; multi-step denotes the need for extensive sequence planning in I-PHYRE, considering each intervention can significantly alter subsequent choices; and in-situ implies the necessity for timely object manipulation within a scene, where minor timing deviations can result in task failure. We formulate four game splits to scrutinize agents' learning and generalization of essential principles of interactive physical reasoning, fostering learning through interaction with representative scenarios. Our exploration involves three planning strategies and examines several supervised and reinforcement agents' zero-shot generalization proficiency on I-PHYRE. The outcomes highlight a notable gap between existing learning algorithms and human performance, emphasizing the imperative for more research in enhancing agents with interactive physical reasoning capabilities. The environment and baselines will be made publicly available.
comment: 21 pages, ICLR 2024
♻ ☆ OpenFMNav: Towards Open-Set Zero-Shot Object Navigation via Vision-Language Foundation Models NAACL 2024
Object navigation (ObjectNav) requires an agent to navigate through unseen environments to find queried objects. Many previous methods attempted to solve this task by relying on supervised or reinforcement learning, where they are trained on limited household datasets with close-set objects. However, two key challenges are unsolved: understanding free-form natural language instructions that demand open-set objects, and generalizing to new environments in a zero-shot manner. Aiming to solve the two challenges, in this paper, we propose OpenFMNav, an Open-set Foundation Model based framework for zero-shot object Navigation. We first unleash the reasoning abilities of large language models (LLMs) to extract proposed objects from natural language instructions that meet the user's demand. We then leverage the generalizability of large vision language models (VLMs) to actively discover and detect candidate objects from the scene, building a Versatile Semantic Score Map (VSSM). Then, by conducting common sense reasoning on VSSM, our method can perform effective language-guided exploration and exploitation of the scene and finally reach the goal. By leveraging the reasoning and generalizing abilities of foundation models, our method can understand free-form human instructions and perform effective open-set zero-shot navigation in diverse environments. Extensive experiments on the HM3D ObjectNav benchmark show that our method surpasses all the strong baselines on all metrics, proving our method's effectiveness. Furthermore, we perform real robot demonstrations to validate our method's open-set-ness and generalizability to real-world environments.
comment: NAACL 2024 Findings
♻ ☆ Towards Massive Interaction with Generalist Robotics: A Systematic Review of XR-enabled Remote Human-Robot Interaction Systems
This survey provides an exhaustive review of the applications of extended reality (XR) technologies in the field of remote human-computer interaction (HRI). We developed a systematic search strategy based on the PRISMA methodology. From the initial 2,561 articles selected, 100 research papers that met our inclusion criteria were included. We categorized and summarized the domain in detail, delving into XR technologies, including augmented reality (AR), virtual reality (VR), and mixed reality (MR), and their applications in facilitating intuitive and effective remote control and interaction with robotic systems.The survey highlights existing articles on the application of XR technologies, user experience enhancement, and various interaction designs for XR in remote HRI, providing insights into current trends and future directions. We also identified potential gaps and opportunities for future research to improve remote HRI systems through XR technology to guide and inform future XR and robotics research.
♻ ☆ Directionality-Aware Mixture Model Parallel Sampling for Efficient Linear Parameter Varying Dynamical System Learning
The Linear Parameter Varying Dynamical System (LPV-DS) is an effective approach that learns stable, time-invariant motion policies using statistical modeling and semi-definite optimization to encode complex motions for reactive robot control. Despite its strengths, the LPV-DS learning approach faces challenges in achieving a high model accuracy without compromising the computational efficiency. To address this, we introduce the Directionality-Aware Mixture Model (DAMM), a novel statistical model that applies the Riemannian metric on the n-sphere $\mathbb{S}^n$ to efficiently blend non-Euclidean directional data with $\mathbb{R}^m$ Euclidean states. Additionally, we develop a hybrid Markov chain Monte Carlo technique that combines Gibbs Sampling with Split/Merge Proposal, allowing for parallel computation to drastically speed up inference. Our extensive empirical tests demonstrate that LPV-DS integrated with DAMM achieves higher reproduction accuracy, better model efficiency, and near real-time/online learning compared to standard estimation methods on various datasets. Lastly, we demonstrate its suitability for incrementally learning multi-behavior policies in real-world robot experiments.
♻ ☆ DTG : Diffusion-based Trajectory Generation for Mapless Global Navigation
We present a novel end-to-end diffusion-based trajectory generation method, DTG, for mapless global navigation in challenging outdoor scenarios with occlusions and unstructured off-road features like grass, buildings, bushes, etc. Given a distant goal, our approach computes a trajectory that satisfies the following goals: (1) minimize the travel distance to the goal; (2) maximize the traversability by choosing paths that do not lie in undesirable areas. Specifically, we present a novel Conditional RNN(CRNN) for diffusion models to efficiently generate trajectories. Furthermore, we propose an adaptive training method that ensures that the diffusion model generates more traversable trajectories. We evaluate our methods in various outdoor scenes and compare the performance with other global navigation algorithms on a Husky robot. In practice, we observe at least a 15% improvement in traveling distance and around a 7% improvement in traversability.
comment: 10 pages
♻ ☆ Fast LiDAR Informed Visual Search in Unseen Indoor Environments
This paper explores the problem of planning for visual search without prior map information. We leverage the pixel-wise environment perception problem where one is given wide Field of View 2D scan data and must perform LiDAR segmentation to contextually label points in the surroundings. These pixel classifications provide an informed prior on which to plan next best viewpoints during visual search tasks. We present LIVES: LiDAR Informed Visual Search, a method aimed at finding objects of interest in unknown indoor environments. A robust map-free classifier is trained from expert data collected using a simple cart platform equipped with a map-based classifier. An autonomous exploration planner takes the contextual data from scans and uses that prior to plan viewpoints more likely to yield detection of the search target. We propose a utility function that accounts for traditional metrics like information gain and path cost and for the contextual information. LIVES is baselined against several existing exploration methods in simulation to verify its performance. It is validated in real-world experiments with single and multiple search objects with a Spot robot in two unseen environments. Videos of experiments, implementation details and open source code can be found at https://sites.google.com/view/lives-2024/home.
comment: 6 pages + references. 6 figures. 1 algorithm. 1 table
♻ ☆ Diagrammatic Instructions to Specify Spatial Objectives and Constraints with Applications to Mobile Base Placement
This paper introduces Spatial Diagrammatic Instructions (SDIs), an approach for human operators to specify objectives and constraints that are related to spatial regions in the working environment. Human operators are enabled to sketch out regions directly on camera images that correspond to the objectives and constraints. These sketches are projected to 3D spatial coordinates, and continuous Spatial Instruction Maps (SIMs) are learned upon them. These maps can then be integrated into optimization problems for tasks of robots. In particular, we demonstrate how Spatial Diagrammatic Instructions can be applied to solve the Base Placement Problem of mobile manipulators, which concerns the best place to put the manipulator to facilitate a certain task. Human operators can specify, via sketch, spatial regions of interest for a manipulation task and permissible regions for the mobile manipulator to be at. Then, an optimization problem that maximizes the manipulator's reachability, or coverage, over the designated regions of interest while remaining in the permissible regions is solved. We provide extensive empirical evaluations, and show that our formulation of Spatial Instruction Maps provides accurate representations of user-specified diagrammatic instructions. Furthermore, we demonstrate that our diagrammatic approach to the Mobile Base Placement Problem enables higher quality solutions and faster run-time.
♻ ☆ Greedy Perspectives: Multi-Drone View Planning for Collaborative Perception in Cluttered Environments IROS'24
Deployment of teams of aerial robots could enable large-scale filming of dynamic groups of people (actors) in complex environments for applications in areas such as team sports and cinematography. Toward this end, methods for submodular maximization via sequential greedy planning can be used for scalable optimization of camera views across teams of robots but face challenges with efficient coordination in cluttered environments. Obstacles can produce occlusions and increase chances of inter-robot collision which can violate requirements for near-optimality guarantees. To coordinate teams of aerial robots in filming groups of people in dense environments, a more general view-planning approach is required. We explore how collision and occlusion impact performance in filming applications through the development of a multi-robot multi-actor view planner with an occlusion-aware objective for filming groups of people and compare with a formation planner and a greedy planner that ignores inter-robot collisions. We evaluate our approach based on five test environments and complex multi-actor behaviors. Compared with a formation planner, our sequential planner generates 14% greater view reward over the actors for three scenarios and comparable performance to formation planning on two others. We also observe near identical view rewards for sequential planning both with and without inter-robot collision constraints which indicates that robots are able to avoid collisions without impairing performance in the perception task. Overall, we demonstrate effective coordination of teams of aerial robots for filming groups that may split, merge, or spread apart and in environments cluttered with obstacles that may cause collisions or occlusions.
comment: Submitted to IROS'24; 8 pages, 8 figures, 2 tables
♻ ☆ Optimizing Crowd-Aware Multi-Agent Path Finding through Local Broadcasting with Graph Neural Networks
Multi-Agent Path Finding (MAPF) in crowded environments presents a challenging problem in motion planning, aiming to find collision-free paths for all agents in the system. MAPF finds a wide range of applications in various domains, including aerial swarms, autonomous warehouse robotics, and self-driving vehicles. Current approaches to MAPF generally fall into two main categories: centralized and decentralized planning. Centralized planning suffers from the curse of dimensionality when the number of agents or states increases and thus does not scale well in large and complex environments. On the other hand, decentralized planning enables agents to engage in real-time path planning within a partially observable environment, demonstrating implicit coordination. However, they suffer from slow convergence and performance degradation in dense environments. In this paper, we introduce CRAMP, a novel crowd-aware decentralized reinforcement learning approach to address this problem by enabling efficient local communication among agents via Graph Neural Networks (GNNs), facilitating situational awareness and decision-making capabilities in congested environments. We test CRAMP on simulated environments and demonstrate that our method outperforms the state-of-the-art decentralized methods for MAPF on various metrics. CRAMP improves the solution quality up to 59% measured in makespan and collision count, and up to 35% improvement in success rate in comparison to previous methods.
comment: 8 pages, 5 figures, 2 tables
♻ ☆ SO(2)-Equivariant Downwash Models for Close Proximity Flight
Multirotors flying in close proximity induce aerodynamic wake effects on each other through propeller downwash. Conventional methods have fallen short of providing adequate 3D force-based models that can be incorporated into robust control paradigms for deploying dense formations. Thus, learning a model for these downwash patterns presents an attractive solution. In this paper, we present a novel learning-based approach for modelling the downwash forces that exploits the latent geometries (i.e. symmetries) present in the problem. We demonstrate that when trained with only 5 minutes of real-world flight data, our geometry-aware model outperforms state-of-the-art baseline models trained with more than 15 minutes of data. In dense real-world flights with two vehicles, deploying our model online improves 3D trajectory tracking by nearly 36% on average (and vertical tracking by 56%).
♻ ☆ Decision-Oriented Learning Using Differentiable Submodular Maximization for Multi-Robot Coordination
We present a differentiable, decision-oriented learning framework for cost prediction in a class of multi-robot decision-making problems, in which the robots need to trade off the task performance with the costs of taking actions when they select actions to take. Specifically, we consider the cases where the task performance is measured by a known monotone submodular function (e.g., coverage, mutual information), and the cost of actions depends on the context (e.g., wind and terrain conditions). We need to learn a function that maps the context to the costs. Classically, we treat such a learning problem and the downstream decision-making problem as two decoupled problems, i.e., we first learn to predict the cost function without considering the downstream decision-making problem, and then use the learned function for predicting the cost and using it in the decision-making problem. However, the loss function used in learning a prediction function may not be aligned with the downstream decision-making. We propose a decision-oriented learning framework that incorporates the downstream task performance in the prediction phase via a differentiable optimization layer. The main computational challenge in such a framework is to make the combinatorial optimization, i.e., non-monotone submodular maximization, differentiable. This function is not naturally differentiable. We propose the Differentiable Cost Scaled Greedy algorithm (D-CSG), which is a continuous and differentiable relaxation of CSG. We demonstrate the efficacy of the proposed framework through numerical simulations. The results show that the proposed framework can result in better performance than the traditional two-stage approach.
comment: arXiv admin note: text overlap with arXiv:2303.01543
♻ ☆ Pre-Trained Masked Image Model for Mobile Robot Navigation ICRA 2024
2D top-down maps are commonly used for the navigation and exploration of mobile robots through unknown areas. Typically, the robot builds the navigation maps incrementally from local observations using onboard sensors. Recent works have shown that predicting the structural patterns in the environment through learning-based approaches can greatly enhance task efficiency. While many such works build task-specific networks using limited datasets, we show that the existing foundational vision networks can accomplish the same without any fine-tuning. Specifically, we use Masked Autoencoders, pre-trained on street images, to present novel applications for field-of-view expansion, single-agent topological exploration, and multi-agent exploration for indoor mapping, across different input modalities. Our work motivates the use of foundational vision models for generalized structure prediction-driven applications, especially in the dearth of training data. For more qualitative results see https://raaslab.org/projects/MIM4Robots.
comment: Accepted at ICRA 2024
♻ ☆ GelLink: A Compact Multi-phalanx Finger with Vision-based Tactile Sensing and Proprioception ICRA 2024
Compared to fully-actuated robotic end-effectors, underactuated ones are generally more adaptive, robust, and cost-effective. However, state estimation for underactuated hands is usually more challenging. Vision-based tactile sensors, like Gelsight, can mitigate this issue by providing high-resolution tactile sensing and accurate proprioceptive sensing. As such, we present GelLink, a compact, underactuated, linkage-driven robotic finger with low-cost, high-resolution vision-based tactile sensing and proprioceptive sensing capabilities. In order to reduce the amount of embedded hardware, i.e. the cameras and motors, we optimize the linkage transmission with a planar linkage mechanism simulator and develop a planar reflection simulator to simplify the tactile sensing hardware. As a result, GelLink only requires one motor to actuate the three phalanges, and one camera to capture tactile signals along the entire finger. Overall, GelLink is a compact robotic finger that shows adaptability and robustness when performing grasping tasks. The integration of vision-based tactile sensors can significantly enhance the capabilities of underactuated fingers and potentially broaden their future usage.
comment: Supplement video: https://www.youtube.com/watch?v=hZwUpAig5C0 . 7 pages, 9 figures. ICRA 2024 (IEEE International Conference on Robotics and Automation)
♻ ☆ C3D: Cascade Control with Change Point Detection and Deep Koopman Learning for Autonomous Surface Vehicles
In this paper, we discuss the development and deployment of a robust autonomous system capable of performing various tasks in the maritime domain under unknown dynamic conditions. We investigate a data-driven approach based on modular design for ease of transfer of autonomy across different maritime surface vessel platforms. The data-driven approach alleviates issues related to a priori identification of system models that may become deficient under evolving system behaviors or shifting, unanticipated, environmental influences. Our proposed learning-based platform comprises a deep Koopman system model and a change point detector that provides guidance on domain shifts prompting relearning under severe exogenous and endogenous perturbations. Motion control of the autonomous system is achieved via an optimal controller design. The Koopman linearized model naturally lends itself to a linear-quadratic regulator (LQR) control design. We propose the C3D control architecture Cascade Control with Change Point Detection and Deep Koopman Learning. The framework is verified in station keeping task on an ASV in both simulation and real experiments. The approach achieved at least 13.9 percent improvement in mean distance error in all test cases compared to the methods that do not consider system changes.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ On the Feedback Law in Stochastic Optimal Nonlinear Control
We consider the problem of nonlinear stochastic optimal control. This problem is thought to be fundamentally intractable owing to Bellman's ``curse of dimensionality". We present a result that shows that repeatedly solving an open-loop deterministic problem from the current state with progressively shorter horizons, similar to Model Predictive Control (MPC), results in a feedback policy that is $O(\epsilon^4)$ near to the true global stochastic optimal policy, \nxx{where $\epsilon$ is a perturbation parameter modulating the noise.} We show that the optimal deterministic feedback problem has a perturbation structure in that higher-order terms of the feedback law do not affect lower-order terms, and that this structure is lost in the optimal stochastic feedback problem. Consequently, solving the Stochastic Dynamic Programming problem is highly susceptible to noise, even when tractable, and in practice, the MPC-type feedback law offers superior performance even for stochastic systems.
comment: arXiv admin note: substantial text overlap with arXiv:2002.10505, arXiv:2002.09478
♻ ☆ Body-mounted MR-conditional Robot for Minimally Invasive Liver Intervention
MR-guided microwave ablation (MWA) has proven effective in treating hepatocellular carcinoma (HCC) with small-sized tumors, but the state-of-the-art technique suffers from sub-optimal workflow due to speed and accuracy of needle placement. This paper presents a compact body-mounted MR-conditional robot that can operate in closed-bore MR scanners for accurate needle guidance. The robotic platform consists of two stacked Cartesian XY stages, each with two degrees of freedom, that facilitate needle guidance. The robot is actuated using 3D-printed pneumatic turbines with MR-conditional bevel gear transmission systems. Pneumatic valves and control mechatronics are located inside the MRI control room and are connected to the robot with pneumatic transmission lines and optical fibers. Free space experiments indicated robot-assisted needle insertion error of 2.6$\pm$1.3 mm at an insertion depth of 80 mm. The MR-guided phantom studies were conducted to verify the MR-conditionality and targeting performance of the robot. Future work will focus on the system optimization and validations in animal trials.
comment: 10 figures
♻ ☆ Mind Meets Robots: A Review of EEG-Based Brain-Robot Interaction Systems
Brain-robot interaction (BRI) empowers individuals to control (semi-)automated machines through their brain activity, either passively or actively. In the past decade, BRI systems have achieved remarkable success, predominantly harnessing electroencephalogram (EEG) signals as the central component. This paper offers an up-to-date and exhaustive examination of 87 curated studies published during the last five years (2018-2023), focusing on identifying the research landscape of EEG-based BRI systems. This review aims to consolidate and underscore methodologies, interaction modes, application contexts, system evaluation, existing challenges, and potential avenues for future investigations in this domain. Based on our analysis, we present a BRI system model with three entities: Brain, Robot, and Interaction, depicting the internal relationships of a BRI system. We especially investigate the essence and principles on interaction modes between human brains and robots, a domain that has not yet been identified anywhere. We then discuss these entities with different dimensions encompassed. Within this model, we scrutinize and classify current research, reveal insights, specify challenges, and provide recommendations for future research trajectories in this field. Meanwhile, we envision our findings offer a design space for future human-robot interaction (HRI) research, informing the creation of efficient BRI frameworks.
♻ ☆ Reinforcement Learning with Options and State Representation
The current thesis aims to explore the reinforcement learning field and build on existing methods to produce improved ones to tackle the problem of learning in high-dimensional and complex environments. It addresses such goals by decomposing learning tasks in a hierarchical fashion known as Hierarchical Reinforcement Learning. We start in the first chapter by getting familiar with the Markov Decision Process framework and presenting some of its recent techniques that the following chapters use. We then proceed to build our Hierarchical Policy learning as an answer to the limitations of a single primitive policy. The hierarchy is composed of a manager agent at the top and employee agents at the lower level. In the last chapter, which is the core of this thesis, we attempt to learn lower-level elements of the hierarchy independently of the manager level in what is known as the "Eigenoption". Based on the graph structure of the environment, Eigenoptions allow us to build agents that are aware of the geometric and dynamic properties of the environment. Their decision-making has a special property: it is invariant to symmetric transformations of the environment, allowing as a consequence to greatly reduce the complexity of the learning task.
comment: Master Thesis 2018, MVA ENS Paris-Saclay, Tokyo RIKEN AIP
♻ ☆ Zero-BEV: Zero-shot Projection of Any First-Person Modality to BEV Maps
Bird's-eye view (BEV) maps are an important geometrically structured representation widely used in robotics, in particular self-driving vehicles and terrestrial robots. Existing algorithms either require depth information for the geometric projection, which is not always reliably available, or are trained end-to-end in a fully supervised way to map visual first-person observations to BEV representation, and are therefore restricted to the output modality they have been trained for. In contrast, we propose a new model capable of performing zero-shot projections of any modality available in a first person view to the corresponding BEV map. This is achieved by disentangling the geometric inverse perspective projection from the modality transformation, eg. RGB to occupancy. The method is general and we showcase experiments projecting to BEV three different modalities: semantic segmentation, motion vectors and object bounding boxes detected in first person. We experimentally show that the model outperforms competing methods, in particular the widely used baseline resorting to monocular depth estimation.
♻ ☆ Accelerating Model Predictive Control for Legged Robots through Distributed Optimization
This paper presents a novel approach to enhance Model Predictive Control (MPC) for legged robots through Distributed Optimization. Our method focuses on decomposing the robot dynamics into smaller, parallelizable subsystems, and utilizing the Alternating Direction Method of Multipliers (ADMM) to ensure consensus among them. Each subsystem is managed by its own Optimal Control Problem, with ADMM facilitating consistency between their optimizations. This approach not only decreases the computational time but also allows for effective scaling with more complex robot configurations, facilitating the integration of additional subsystems such as articulated arms on a quadruped robot. We demonstrate, through numerical evaluations, the convergence of our approach on two systems with increasing complexity. In addition, we showcase that our approach converges towards the same solution when compared to a state-of-the-art centralized whole-body MPC implementation. Moreover, we quantitatively compare the computational efficiency of our method to the centralized approach, revealing up to a 75\% reduction in computational time. Overall, our approach offers a promising avenue for accelerating MPC solutions for legged robots, paving the way for more effective utilization of the computational performance of modern hardware.
♻ ☆ Towards Precise 3D Human Pose Estimation with Multi-Perspective Spatial-Temporal Relational Transformers IJCNN 2024
3D human pose estimation captures the human joint points in three-dimensional space while keeping the depth information and physical structure. That is essential for applications that require precise pose information, such as human-computer interaction, scene understanding, and rehabilitation training. Due to the challenges in data collection, mainstream datasets of 3D human pose estimation are primarily composed of multi-view video data collected in laboratory environments, which contains rich spatial-temporal correlation information besides the image frame content. Given the remarkable self-attention mechanism of transformers, capable of capturing the spatial-temporal correlation from multi-view video datasets, we propose a multi-stage framework for 3D sequence-to-sequence (seq2seq) human pose detection. Firstly, the spatial module represents the human pose feature by intra-image content, while the frame-image relation module extracts temporal relationships and 3D spatial positional relationship features between the multi-perspective images. Secondly, the self-attention mechanism is adopted to eliminate the interference from non-human body parts and reduce computing resources. Our method is evaluated on Human3.6M, a popular 3D human pose detection dataset. Experimental results demonstrate that our approach achieves state-of-the-art performance on this dataset. The source code will be available at https://github.com/WUJINHUAN/3D-human-pose.
comment: Accepted to IJCNN 2024. The source code will be available at https://github.com/WUJINHUAN/3D-human-pose
♻ ☆ Uni-RLHF: Universal Platform and Benchmark Suite for Reinforcement Learning with Diverse Human Feedback ICLR 2024
Reinforcement Learning with Human Feedback (RLHF) has received significant attention for performing tasks without the need for costly manual reward design by aligning human preferences. It is crucial to consider diverse human feedback types and various learning methods in different environments. However, quantifying progress in RLHF with diverse feedback is challenging due to the lack of standardized annotation platforms and widely used unified benchmarks. To bridge this gap, we introduce Uni-RLHF, a comprehensive system implementation tailored for RLHF. It aims to provide a complete workflow from real human feedback, fostering progress in the development of practical problems. Uni-RLHF contains three packages: 1) a universal multi-feedback annotation platform, 2) large-scale crowdsourced feedback datasets, and 3) modular offline RLHF baseline implementations. Uni-RLHF develops a user-friendly annotation interface tailored to various feedback types, compatible with a wide range of mainstream RL environments. We then establish a systematic pipeline of crowdsourced annotations, resulting in large-scale annotated datasets comprising more than 15 million steps across 30+ popular tasks. Through extensive experiments, the results in the collected datasets demonstrate competitive performance compared to those from well-designed manual rewards. We evaluate various design choices and offer insights into their strengths and potential areas of improvement. We wish to build valuable open-source platforms, datasets, and baselines to facilitate the development of more robust and reliable RLHF solutions based on realistic human feedback. The website is available at https://uni-rlhf.github.io/.
comment: Published as a conference paper at ICLR 2024. The website is available at https://uni-rlhf.github.io/
♻ ☆ Vehicle Trajectory Tracking Through Magnetic Sensors: A Case Study of Two-lane Road
Intelligent Transportation Systems (ITS) have a pressing need for efficient and reliable traffic surveillance solutions. This paper for the first time proposes a surveillance system that utilizes low-cost magnetic sensors for detecting and tracking vehicles continuously along the road. The system uses multiple sensors mounted along the roadside and lane boundaries to capture the movement of vehicles. Real-time measurement data is collected by base stations and processed to produce vehicle trajectories that include position, timestamp, and speed. To address the challenge of tracking vehicles continuously on a road network using a large amount of unlabeled magnetic sensor measurements, we first define a vehicle trajectory tracking problem. We then propose a graph-based data association algorithm to track each detected vehicle, and design a related online algorithm framework respectively. We finally validate the performance via both experimental simulation and real-world road deployment. The experimental results demonstrate that the proposed solution provides a cost-effective solution to capture the driving status of vehicles and on that basis form various traffic safety and efficiency applications.
Artificial Intelligence 75
☆ Enabling Uncertainty Estimation in Iterative Neural Networks
Turning pass-through network architectures into iterative ones, which use their own output as input, is a well-known approach for boosting performance. In this paper, we argue that such architectures offer an additional benefit: The convergence rate of their successive outputs is highly correlated with the accuracy of the value to which they converge. Thus, we can use the convergence rate as a useful proxy for uncertainty. This results in an approach to uncertainty estimation that provides state-of-the-art estimates at a much lower computational cost than techniques like Ensembles, and without requiring any modifications to the original iterative model. We demonstrate its practical value by embedding it in two application domains: road detection in aerial images and the estimation of aerodynamic properties of 2D and 3D shapes.
☆ Improving Diffusion Models's Data-Corruption Resistance using Scheduled Pseudo-Huber Loss
Diffusion models are known to be vulnerable to outliers in training data. In this paper we study an alternative diffusion loss function, which can preserve the high quality of generated data like the original squared $L_{2}$ loss while at the same time being robust to outliers. We propose to use pseudo-Huber loss function with a time-dependent parameter to allow for the trade-off between robustness on the most vulnerable early reverse-diffusion steps and fine details restoration on the final steps. We show that pseudo-Huber loss with the time-dependent parameter exhibits better performance on corrupted datasets in both image and audio domains. In addition, the loss function we propose can potentially help diffusion models to resist dataset corruption while not requiring data filtering or purification compared to conventional training algorithms.
comment: 13 pages, 16 figures
☆ Towards a Formalisation of Value-based Actions and Consequentialist Ethics
Agents act to bring about a state of the world that is more compatible with their personal or institutional values. To formalise this intuition, the paper proposes an action framework based on the STRIPS formalisation. Technically, the contribution expresses actions in terms of Value-based Formal Reasoning (VFR), which provides a set of propositions derived from an Agent's value profile and the Agent's assessment of propositions with respect to the profile. Conceptually, the contribution provides a computational framework for a form of consequentialist ethics which is satisficing, luralistic, act-based, and preferential.
☆ One-Shot Domain Incremental Learning IJCNN
Domain incremental learning (DIL) has been discussed in previous studies on deep neural network models for classification. In DIL, we assume that samples on new domains are observed over time. The models must classify inputs on all domains. In practice, however, we may encounter a situation where we need to perform DIL under the constraint that the samples on the new domain are observed only infrequently. Therefore, in this study, we consider the extreme case where we have only one sample from the new domain, which we call one-shot DIL. We first empirically show that existing DIL methods do not work well in one-shot DIL. We have analyzed the reason for this failure through various investigations. According to our analysis, we clarify that the difficulty of one-shot DIL is caused by the statistics in the batch normalization layers. Therefore, we propose a technique regarding these statistics and demonstrate the effectiveness of our technique through experiments on open datasets.
comment: accepted at IEEE International Joint Conference on Neural Networks (IJCNN) 2024
☆ Investigation of the effectiveness of applying ChatGPT in Dialogic Teaching Using Electroencephalography
In recent years, the rapid development of artificial intelligence technology, especially the emergence of large language models (LLMs) such as ChatGPT, has presented significant prospects for application in the field of education. LLMs possess the capability to interpret knowledge, answer questions, and consider context, thus providing support for dialogic teaching to students. Therefore, an examination of the capacity of LLMs to effectively fulfill instructional roles, thereby facilitating student learning akin to human educators within dialogic teaching scenarios, is an exceptionally valuable research topic. This research recruited 34 undergraduate students as participants, who were randomly divided into two groups. The experimental group engaged in dialogic teaching using ChatGPT, while the control group interacted with human teachers. Both groups learned the histogram equalization unit in the information-related course "Digital Image Processing". The research findings show comparable scores between the two groups on the retention test. However, students who engaged in dialogue with ChatGPT exhibited lower performance on the transfer test. Electroencephalography data revealed that students who interacted with ChatGPT exhibited higher levels of cognitive activity, suggesting that ChatGPT could help students establish a knowledge foundation and stimulate cognitive activity. However, its strengths on promoting students. knowledge application and creativity were insignificant. Based upon the research findings, it is evident that ChatGPT cannot fully excel in fulfilling teaching tasks in the dialogue teaching in information related courses. Combining ChatGPT with traditional human teachers might be a more ideal approach. The synergistic use of both can provide students with more comprehensive learning support, thus contributing to enhancing the quality of teaching.
☆ Understanding the Functional Roles of Modelling Components in Spiking Neural Networks
Spiking neural networks (SNNs), inspired by the neural circuits of the brain, are promising in achieving high computational efficiency with biological fidelity. Nevertheless, it is quite difficult to optimize SNNs because the functional roles of their modelling components remain unclear. By designing and evaluating several variants of the classic model, we systematically investigate the functional roles of key modelling components, leakage, reset, and recurrence, in leaky integrate-and-fire (LIF) based SNNs. Through extensive experiments, we demonstrate how these components influence the accuracy, generalization, and robustness of SNNs. Specifically, we find that the leakage plays a crucial role in balancing memory retention and robustness, the reset mechanism is essential for uninterrupted temporal processing and computational efficiency, and the recurrence enriches the capability to model complex dynamics at a cost of robustness degradation. With these interesting observations, we provide optimization suggestions for enhancing the performance of SNNs in different scenarios. This work deepens the understanding of how SNNs work, which offers valuable guidance for the development of more effective and robust neuromorphic models.
☆ Deep Reinforcement Learning and Mean-Variance Strategies for Responsible Portfolio Optimization AAAI 2024
Portfolio optimization involves determining the optimal allocation of portfolio assets in order to maximize a given investment objective. Traditionally, some form of mean-variance optimization is used with the aim of maximizing returns while minimizing risk, however, more recently, deep reinforcement learning formulations have been explored. Increasingly, investors have demonstrated an interest in incorporating ESG objectives when making investment decisions, and modifications to the classical mean-variance optimization framework have been developed. In this work, we study the use of deep reinforcement learning for responsible portfolio optimization, by incorporating ESG states and objectives, and provide comparisons against modified mean-variance approaches. Our results show that deep reinforcement learning policies can provide competitive performance against mean-variance approaches for responsible portfolio allocation across additive and multiplicative utility functions of financial and ESG responsibility objectives.
comment: Presented at the AAAI 2024 Workshop on AI in Finance for Social Impact
☆ Revisiting the Sleeping Beauty problem
The Sleeping Beauty problem is a probability riddle with no definite solution for more than two decades and its solution is of great interest in many fields of knowledge. There are two main competing solutions to the problem: the halfer approach, and the thirder approach. The main reason for disagreement in the literature is connected to the use of different probability spaces to represent the same probabilistic riddle. In this work, we analyse the problem from a mathematical perspective, identifying probability distributions induced directly from the thought experiment's rules. The precise choices of probability spaces provide both halfer and thirder solutions to the problem. To try and decide on which approach to follow, a criterion involving the information available to Sleeping Beauty is proposed.
comment: 14 pages, 1 figure
☆ CLHA: A Simple yet Effective Contrastive Learning Framework for Human Alignment
Reinforcement learning from human feedback (RLHF) is a crucial technique in aligning large language models (LLMs) with human preferences, ensuring these LLMs behave in beneficial and comprehensible ways to users. However, a longstanding challenge in human alignment techniques based on reinforcement learning lies in their inherent complexity and difficulty in training. To address this challenge, we present a simple yet effective Contrastive Learning Framework for Human Alignment (CLHA) to align LLMs with human preferences directly. CLHA employs a novel rescoring strategy to evaluate the noise within the data by considering its inherent quality and dynamically adjusting the training process. Simultaneously, CLHA utilizes pairwise contrastive loss and adaptive supervised fine-tuning loss to adaptively modify the likelihood of generating responses, ensuring enhanced alignment with human preferences. Using advanced methods, CLHA surpasses other algorithms, showcasing superior performance in terms of reward model scores, automatic evaluations, and human assessments on the widely used ``\textit{Helpful and Harmless}'' dataset.
☆ Deciphering the Interplay between Local Differential Privacy, Average Bayesian Privacy, and Maximum Bayesian Privacy
The swift evolution of machine learning has led to emergence of various definitions of privacy due to the threats it poses to privacy, including the concept of local differential privacy (LDP). Although widely embraced and utilized across numerous domains, this conventional approach to measure privacy still exhibits certain limitations, spanning from failure to prevent inferential disclosure to lack of consideration for the adversary's background knowledge. In this comprehensive study, we introduce Bayesian privacy and delve into the intricate relationship between local differential privacy and its Bayesian counterparts, unveiling novel insights into utility-privacy trade-offs. We introduce a framework that encapsulates both attack and defense strategies, highlighting their interplay and effectiveness. Our theoretical contributions are anchored in the rigorous definitions and relationships between Average Bayesian Privacy (ABP) and Maximum Bayesian Privacy (MBP), encapsulated by equations $\epsilon_{p,a} \leq \frac{1}{\sqrt{2}}\sqrt{(\epsilon_{p,m} + \epsilon)\cdot(e^{\epsilon_{p,m} + \epsilon} - 1)}$ and the equivalence between $\xi$-MBP and $2\xi$-LDP established under uniform prior distribution. These relationships fortify our understanding of the privacy guarantees provided by various mechanisms, leading to the realization that a mechanism satisfying $\xi$-LDP also confers $\xi$-MBP, and vice versa. Our work not only lays the groundwork for future empirical exploration but also promises to enhance the design of privacy-preserving algorithms that do not compromise on utility, thereby fostering the development of trustworthy machine learning solutions.
☆ In the Search for Optimal Multi-view Learning Models for Crop Classification with Global Remote Sensing Data
Crop classification is of critical importance due to its role in studying crop pattern changes, resource management, and carbon sequestration. When employing data-driven techniques for its prediction, utilizing various temporal data sources is necessary. Deep learning models have proven to be effective for this task by mapping time series data to high-level representation for prediction. However, they face substantial challenges when dealing with multiple input patterns. The literature offers limited guidance for Multi-View Learning (MVL) scenarios, as it has primarily focused on exploring fusion strategies with specific encoders and validating them in local regions. In contrast, we investigate the impact of simultaneous selection of the fusion strategy and the encoder architecture evaluated on a global-scale cropland and crop-type classifications. We use a range of five fusion strategies (Input, Feature, Decision, Ensemble, Hybrid) and five temporal encoder architectures (LSTM, GRU, TempCNN, TAE, L-TAE) as possible MVL model configurations. The validation is on the CropHarvest dataset that provides optical, radar, and weather time series, and topographic information as input data. We found that in scenarios with a limited number of labeled samples, a unique configuration is insufficient for all the cases. Instead, a specialized combination, including encoder and fusion strategy, should be meticulously sought. To streamline this search process, we suggest initially identifying the optimal encoder architecture tailored for a particular fusion strategy, and then determining the most suitable fusion strategy for the classification task. We provide a technical framework for researchers exploring crop classification or related tasks through a MVL approach.
comment: submitted to journal
☆ SegICL: A Universal In-context Learning Framework for Enhanced Segmentation in Medical Imaging
Medical image segmentation models adapting to new tasks in a training-free manner through in-context learning is an exciting advancement. Universal segmentation models aim to generalize across the diverse modality of medical images, yet their effectiveness often diminishes when applied to out-of-distribution (OOD) data modalities and tasks, requiring intricate fine-tuning of model for optimal performance. For addressing this challenge, we introduce SegICL, a novel approach leveraging In-Context Learning (ICL) for image segmentation. Unlike existing methods, SegICL has the capability to employ text-guided segmentation and conduct in-context learning with a small set of image-mask pairs, eliminating the need for training the model from scratch or fine-tuning for OOD tasks (including OOD modality and dataset). Extensive experimental validation of SegICL demonstrates a positive correlation between the number of prompt samples and segmentation performance on OOD modalities and tasks. This indicates that SegICL effectively address new segmentation tasks based on contextual information. Additionally, SegICL also exhibits comparable segmentation performance to mainstream models on OOD and in-distribution tasks. Our code will be released soon.
☆ NSINA: A News Corpus for Sinhala LREC
The introduction of large language models (LLMs) has advanced natural language processing (NLP), but their effectiveness is largely dependent on pre-training resources. This is especially evident in low-resource languages, such as Sinhala, which face two primary challenges: the lack of substantial training data and limited benchmarking datasets. In response, this study introduces NSINA, a comprehensive news corpus of over 500,000 articles from popular Sinhala news websites, along with three NLP tasks: news media identification, news category prediction, and news headline generation. The release of NSINA aims to provide a solution to challenges in adapting LLMs to Sinhala, offering valuable resources and benchmarks for improving NLP in the Sinhala language. NSINA is the largest news corpus for Sinhala, available up to date.
comment: Accepted to LREC-COLING 2024 (The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation)
☆ FedFixer: Mitigating Heterogeneous Label Noise in Federated Learning
Federated Learning (FL) heavily depends on label quality for its performance. However, the label distribution among individual clients is always both noisy and heterogeneous. The high loss incurred by client-specific samples in heterogeneous label noise poses challenges for distinguishing between client-specific and noisy label samples, impacting the effectiveness of existing label noise learning approaches. To tackle this issue, we propose FedFixer, where the personalized model is introduced to cooperate with the global model to effectively select clean client-specific samples. In the dual models, updating the personalized model solely at a local level can lead to overfitting on noisy data due to limited samples, consequently affecting both the local and global models' performance. To mitigate overfitting, we address this concern from two perspectives. Firstly, we employ a confidence regularizer to alleviate the impact of unconfident predictions caused by label noise. Secondly, a distance regularizer is implemented to constrain the disparity between the personalized and global models. We validate the effectiveness of FedFixer through extensive experiments on benchmark datasets. The results demonstrate that FedFixer can perform well in filtering noisy label samples on different clients, especially in highly heterogeneous label noise scenarios.
comment: accepted by AAA24
☆ PE: A Poincare Explanation Method for Fast Text Hierarchy Generation
The black-box nature of deep learning models in NLP hinders their widespread application. The research focus has shifted to Hierarchical Attribution (HA) for its ability to model feature interactions. Recent works model non-contiguous combinations with a time-costly greedy search in Eculidean spaces, neglecting underlying linguistic information in feature representations. In this work, we introduce a novel method, namely Poincar\'e Explanation (PE), for modeling feature interactions using hyperbolic spaces in an $O(n^2logn)$ time complexity. Inspired by Poincar\'e model, we propose a framework to project the embeddings into hyperbolic spaces, which exhibit better inductive biases for syntax and semantic hierarchical structures. Eventually, we prove that the hierarchical clustering process in the projected space could be viewed as building a minimum spanning tree and propose a time efficient algorithm. Experimental results demonstrate the effectiveness of our approach.
comment: 12 pages, 10 figures
☆ QKFormer: Hierarchical Spiking Transformer using Q-K Attention
Spiking Transformers, which integrate Spiking Neural Networks (SNNs) with Transformer architectures, have attracted significant attention due to their potential for energy efficiency and high performance. However, existing models in this domain still suffer from suboptimal performance. We introduce several innovations to improve the performance: i) We propose a novel spike-form Q-K attention mechanism, tailored for SNNs, which efficiently models the importance of token or channel dimensions through binary vectors with linear complexity. ii) We incorporate the hierarchical structure, which significantly benefits the performance of both the brain and artificial neural networks, into spiking transformers to obtain multi-scale spiking representation. iii) We design a versatile and powerful patch embedding module with a deformed shortcut specifically for spiking transformers. Together, we develop QKFormer, a hierarchical spiking transformer based on Q-K attention with direct training. QKFormer shows significantly superior performance over existing state-of-the-art SNN models on various mainstream datasets. Notably, with comparable size to Spikformer (66.34 M, 74.81%), QKFormer (64.96 M) achieves a groundbreaking top-1 accuracy of 85.65% on ImageNet-1k, substantially outperforming Spikformer by 10.84%. To our best knowledge, this is the first time that directly training SNNs have exceeded 85% accuracy on ImageNet-1K. The code and models are publicly available at https://github.com/zhouchenlin2096/QKFormer
comment: 10 pages, code: https://github.com/zhouchenlin2096/QKFormer
☆ Efficient Information Extraction in Few-Shot Relation Classification through Contrastive Representation Learning NAACL 2024
Differentiating relationships between entity pairs with limited labeled instances poses a significant challenge in few-shot relation classification. Representations of textual data extract rich information spanning the domain, entities, and relations. In this paper, we introduce a novel approach to enhance information extraction combining multiple sentence representations and contrastive learning. While representations in relation classification are commonly extracted using entity marker tokens, we argue that substantial information within the internal model representations remains untapped. To address this, we propose aligning multiple sentence representations, such as the [CLS] token, the [MASK] token used in prompting, and entity marker tokens. Our method employs contrastive learning to extract complementary discriminative information from these individual representations. This is particularly relevant in low-resource settings where information is scarce. Leveraging multiple sentence representations is especially effective in distilling discriminative information for relation classification when additional information, like relation descriptions, are not available. We validate the adaptability of our approach, maintaining robust performance in scenarios that include relation descriptions, and showcasing its flexibility to adapt to different resource constraints.
comment: NAACL 2024
☆ An Intermediate Fusion ViT Enables Efficient Text-Image Alignment in Diffusion Models
Diffusion models have been widely used for conditional data cross-modal generation tasks such as text-to-image and text-to-video. However, state-of-the-art models still fail to align the generated visual concepts with high-level semantics in a language such as object count, spatial relationship, etc. We approach this problem from a multimodal data fusion perspective and investigate how different fusion strategies can affect vision-language alignment. We discover that compared to the widely used early fusion of conditioning text in a pretrained image feature space, a specially designed intermediate fusion can: (i) boost text-to-image alignment with improved generation quality and (ii) improve training and inference efficiency by reducing low-rank text-to-image attention calculations. We perform experiments using a text-to-image generation task on the MS-COCO dataset. We compare our intermediate fusion mechanism with the classic early fusion mechanism on two common conditioning methods on a U-shaped ViT backbone. Our intermediate fusion model achieves a higher CLIP Score and lower FID, with 20% reduced FLOPs, and 50% increased training speed compared to a strong U-ViT baseline with an early fusion.
☆ Hallucination Detection in Foundation Models for Decision-Making: A Flexible Definition and Review of the State of the Art
Autonomous systems are soon to be ubiquitous, from manufacturing autonomy to agricultural field robots, and from health care assistants to the entertainment industry. The majority of these systems are developed with modular sub-components for decision-making, planning, and control that may be hand-engineered or learning-based. While these existing approaches have been shown to perform well under the situations they were specifically designed for, they can perform especially poorly in rare, out-of-distribution scenarios that will undoubtedly arise at test-time. The rise of foundation models trained on multiple tasks with impressively large datasets from a variety of fields has led researchers to believe that these models may provide common sense reasoning that existing planners are missing. Researchers posit that this common sense reasoning will bridge the gap between algorithm development and deployment to out-of-distribution tasks, like how humans adapt to unexpected scenarios. Large language models have already penetrated the robotics and autonomous systems domains as researchers are scrambling to showcase their potential use cases in deployment. While this application direction is very promising empirically, foundation models are known to hallucinate and generate decisions that may sound reasonable, but are in fact poor. We argue there is a need to step back and simultaneously design systems that can quantify the certainty of a model's decision, and detect when it may be hallucinating. In this work, we discuss the current use cases of foundation models for decision-making tasks, provide a general definition for hallucinations with examples, discuss existing approaches to hallucination detection and mitigation with a focus on decision problems, and explore areas for further research in this exciting field.
comment: 31 pages, 2 tables
☆ Harnessing the power of LLMs for normative reasoning in MASs AAMAS 2024
Software agents, both human and computational, do not exist in isolation and often need to collaborate or coordinate with others to achieve their goals. In human society, social mechanisms such as norms ensure efficient functioning, and these techniques have been adopted by researchers in multi-agent systems (MAS) to create socially aware agents. However, traditional techniques have limitations, such as operating in limited environments often using brittle symbolic reasoning. The advent of Large Language Models (LLMs) offers a promising solution, providing a rich and expressive vocabulary for norms and enabling norm-capable agents that can perform a range of tasks such as norm discovery, normative reasoning and decision-making. This paper examines the potential of LLM-based agents to acquire normative capabilities, drawing on recent Natural Language Processing (NLP) and LLM research. We present our vision for creating normative LLM agents. In particular, we discuss how the recently proposed "LLM agent" approaches can be extended to implement such normative LLM agents. We also highlight challenges in this emerging field. This paper thus aims to foster collaboration between MAS, NLP and LLM researchers in order to advance the field of normative agents.
comment: 12 pages, 1 figure, accepted to COINE 2024 workshop at AAMAS 2024 (https://coin-workshop.github.io/coine-2024-auckland/accepted_papers.html)
☆ Causal Discovery from Poisson Branching Structural Causal Model Using High-Order Cumulant with Path Analysis AAAI-2024
Count data naturally arise in many fields, such as finance, neuroscience, and epidemiology, and discovering causal structure among count data is a crucial task in various scientific and industrial scenarios. One of the most common characteristics of count data is the inherent branching structure described by a binomial thinning operator and an independent Poisson distribution that captures both branching and noise. For instance, in a population count scenario, mortality and immigration contribute to the count, where survival follows a Bernoulli distribution, and immigration follows a Poisson distribution. However, causal discovery from such data is challenging due to the non-identifiability issue: a single causal pair is Markov equivalent, i.e., $X\rightarrow Y$ and $Y\rightarrow X$ are distributed equivalent. Fortunately, in this work, we found that the causal order from $X$ to its child $Y$ is identifiable if $X$ is a root vertex and has at least two directed paths to $Y$, or the ancestor of $X$ with the most directed path to $X$ has a directed path to $Y$ without passing $X$. Specifically, we propose a Poisson Branching Structure Causal Model (PB-SCM) and perform a path analysis on PB-SCM using high-order cumulants. Theoretical results establish the connection between the path and cumulant and demonstrate that the path information can be obtained from the cumulant. With the path information, causal order is identifiable under some graphical conditions. A practical algorithm for learning causal structure under PB-SCM is proposed and the experiments demonstrate and verify the effectiveness of the proposed method.
comment: Accepted by AAAI-2024
☆ LLMs Are Few-Shot In-Context Low-Resource Language Learners
In-context learning (ICL) empowers large language models (LLMs) to perform diverse tasks in underrepresented languages using only short in-context information, offering a crucial avenue for narrowing the gap between high-resource and low-resource languages. Nonetheless, there is only a handful of works explored ICL for low-resource languages with most of them focusing on relatively high-resource languages, such as French and Spanish. In this work, we extensively study ICL and its cross-lingual variation (X-ICL) on 25 low-resource and 7 relatively higher-resource languages. Our study not only assesses the effectiveness of ICL with LLMs in low-resource languages but also identifies the shortcomings of in-context label alignment, and introduces a more effective alternative: query alignment. Moreover, we provide valuable insights into various facets of ICL for low-resource languages. Our study concludes the significance of few-shot in-context information on enhancing the low-resource understanding quality of LLMs through semantically relevant information by closing the language gap in the target language and aligning the semantics between the targeted low-resource and the high-resource language that the model is proficient in. Our work highlights the importance of advancing ICL research, particularly for low-resource languages.
☆ Return to Tradition: Learning Reliable Heuristics with Classical Machine Learning ICAPS 2024
Current approaches for learning for planning have yet to achieve competitive performance against classical planners in several domains, and have poor overall performance. In this work, we construct novel graph representations of lifted planning tasks and use the WL algorithm to generate features from them. These features are used with classical machine learning methods which have up to 2 orders of magnitude fewer parameters and train up to 3 orders of magnitude faster than the state-of-the-art deep learning for planning models. Our novel approach, WL-GOOSE, reliably learns heuristics from scratch and outperforms the $h^{\text{FF}}$ heuristic in a fair competition setting. It also outperforms or ties with LAMA on 4 out of 10 domains on coverage and 7 out of 10 domains on plan quality. WL-GOOSE is the first learning for planning model which achieves these feats. Furthermore, we study the connections between our novel WL feature generation method, previous theoretically flavoured learning architectures, and Description Logic Features for planning.
comment: Extended version of ICAPS 2024 paper
☆ Learning To Guide Human Decision Makers With Vision-Language Models
There is increasing interest in developing AIs for assisting human decision making in \textit{high-stakes} tasks, such as medical diagnosis, for the purpose of improving decision quality and reducing cognitive strain. % Mainstream approaches team up an expert with a machine learning model to which safer decisions are offloaded, thus letting the former focus on cases that demand their attention. % This \textit{separation of responsibilities} setup, however, is inadequate for high-stakes scenarios. On the one hand, the expert may end up over-relying on the machine's decisions due to \textit{anchoring bias}, thus losing the human oversight that is increasingly being required by regulatory agencies to ensure trustworthy AI. On the other hand, the expert is left entirely unassisted on the (typically hardest) decisions on which the model abstained. % As a remedy, we introduce \textit{learning to guide} (LTG), an alternative framework in which -- rather than taking control from the human expert -- the machine provides \textit{guidance} useful for decision making, and the human is entirely responsible for coming up with a decision. % In order to ensure guidance is \textit{interpretable} and \textit{task-specific}, we develop \method, an approach for turning \textit{any} vision-language model into a capable generator of textual guidance by leveraging a modicum of human feedback. % Our empirical evaluation highlights the promise of \method on a challenging, real-world medical diagnosis task.
☆ LSTTN: A Long-Short Term Transformer-based Spatio-temporal Neural Network for Traffic Flow Forecasting
Accurate traffic forecasting is a fundamental problem in intelligent transportation systems and learning long-range traffic representations with key information through spatiotemporal graph neural networks (STGNNs) is a basic assumption of current traffic flow prediction models. However, due to structural limitations, existing STGNNs can only utilize short-range traffic flow data; therefore, the models cannot adequately learn the complex trends and periodic features in traffic flow. Besides, it is challenging to extract the key temporal information from the long historical traffic series and obtain a compact representation. To solve the above problems, we propose a novel LSTTN (Long-Short Term Transformer-based Network) framework comprehensively considering the long- and short-term features in historical traffic flow. First, we employ a masked subseries Transformer to infer the content of masked subseries from a small portion of unmasked subseries and their temporal context in a pretraining manner, forcing the model to efficiently learn compressed and contextual subseries temporal representations from long historical series. Then, based on the learned representations, long-term trend is extracted by using stacked 1D dilated convolution layers, and periodic features are extracted by dynamic graph convolution layers. For the difficulties in making time-step level prediction, LSTTN adopts a short-term trend extractor to learn fine-grained short-term temporal features. Finally, LSTTN fuses the long-term trend, periodic features and short-term features to obtain the prediction results. Experiments on four real-world datasets show that in 60-minute-ahead long-term forecasting, the LSTTN model achieves a minimum improvement of 5.63\% and a maximum improvement of 16.78\% over baseline models. The source code is available at https://github.com/GeoX-Lab/LSTTN.
comment: 15 pages, 10 figures, 6 tables
☆ FedAC: A Adaptive Clustered Federated Learning Framework for Heterogeneous Data
Clustered federated learning (CFL) is proposed to mitigate the performance deterioration stemming from data heterogeneity in federated learning (FL) by grouping similar clients for cluster-wise model training. However, current CFL methods struggle due to inadequate integration of global and intra-cluster knowledge and the absence of an efficient online model similarity metric, while treating the cluster count as a fixed hyperparameter limits flexibility and robustness. In this paper, we propose an adaptive CFL framework, named FedAC, which (1) efficiently integrates global knowledge into intra-cluster learning by decoupling neural networks and utilizing distinct aggregation methods for each submodule, significantly enhancing performance; (2) includes a costeffective online model similarity metric based on dimensionality reduction; (3) incorporates a cluster number fine-tuning module for improved adaptability and scalability in complex, heterogeneous environments. Extensive experiments show that FedAC achieves superior empirical performance, increasing the test accuracy by around 1.82% and 12.67% on CIFAR-10 and CIFAR-100 datasets, respectively, under different non-IID settings compared to SOTA methods.
comment: 14 pages, 4 figures
☆ DeepMachining: Online Prediction of Machining Errors of Lathe Machines
We describe DeepMachining, a deep learning-based AI system for online prediction of machining errors of lathe machine operations. We have built and evaluated DeepMachining based on manufacturing data from factories. Specifically, we first pretrain a deep learning model for a given lathe machine's operations to learn the salient features of machining states. Then, we fine-tune the pretrained model to adapt to specific machining tasks. We demonstrate that DeepMachining achieves high prediction accuracy for multiple tasks that involve different workpieces and cutting tools. To the best of our knowledge, this work is one of the first factory experiments using pre-trained deep-learning models to predict machining errors of lathe machines.
☆ CodeS: Natural Language to Code Repository via Multi-Layer Sketch
The impressive performance of large language models (LLMs) on code-related tasks has shown the potential of fully automated software development. In light of this, we introduce a new software engineering task, namely Natural Language to code Repository (NL2Repo). This task aims to generate an entire code repository from its natural language requirements. To address this task, we propose a simple yet effective framework CodeS, which decomposes NL2Repo into multiple sub-tasks by a multi-layer sketch. Specifically, CodeS includes three modules: RepoSketcher, FileSketcher, and SketchFiller. RepoSketcher first generates a repository's directory structure for given requirements; FileSketcher then generates a file sketch for each file in the generated structure; SketchFiller finally fills in the details for each function in the generated file sketch. To rigorously assess CodeS on the NL2Repo task, we carry out evaluations through both automated benchmarking and manual feedback analysis. For benchmark-based evaluation, we craft a repository-oriented benchmark, SketchEval, and design an evaluation metric, SketchBLEU. For feedback-based evaluation, we develop a VSCode plugin for CodeS and engage 30 participants in conducting empirical studies. Extensive experiments prove the effectiveness and practicality of CodeS on the NL2Repo task.
comment: https://github.com/NL2Code/CodeS
☆ $\textit{LinkPrompt}$: Natural and Universal Adversarial Attacks on Prompt-based Language Models NAACL2024
Prompt-based learning is a new language model training paradigm that adapts the Pre-trained Language Models (PLMs) to downstream tasks, which revitalizes the performance benchmarks across various natural language processing (NLP) tasks. Instead of using a fixed prompt template to fine-tune the model, some research demonstrates the effectiveness of searching for the prompt via optimization. Such prompt optimization process of prompt-based learning on PLMs also gives insight into generating adversarial prompts to mislead the model, raising concerns about the adversarial vulnerability of this paradigm. Recent studies have shown that universal adversarial triggers (UATs) can be generated to alter not only the predictions of the target PLMs but also the prediction of corresponding Prompt-based Fine-tuning Models (PFMs) under the prompt-based learning paradigm. However, UATs found in previous works are often unreadable tokens or characters and can be easily distinguished from natural texts with adaptive defenses. In this work, we consider the naturalness of the UATs and develop $\textit{LinkPrompt}$, an adversarial attack algorithm to generate UATs by a gradient-based beam search algorithm that not only effectively attacks the target PLMs and PFMs but also maintains the naturalness among the trigger tokens. Extensive results demonstrate the effectiveness of $\textit{LinkPrompt}$, as well as the transferability of UATs generated by \textit{LinkPrompt} to open-sourced Large Language Model (LLM) Llama2 and API-accessed LLM GPT-3.5-turbo.
comment: Accepted to the main conference of NAACL2024
☆ DOCTR: Disentangled Object-Centric Transformer for Point Scene Understanding
Point scene understanding is a challenging task to process real-world scene point cloud, which aims at segmenting each object, estimating its pose, and reconstructing its mesh simultaneously. Recent state-of-the-art method first segments each object and then processes them independently with multiple stages for the different sub-tasks. This leads to a complex pipeline to optimize and makes it hard to leverage the relationship constraints between multiple objects. In this work, we propose a novel Disentangled Object-Centric TRansformer (DOCTR) that explores object-centric representation to facilitate learning with multiple objects for the multiple sub-tasks in a unified manner. Each object is represented as a query, and a Transformer decoder is adapted to iteratively optimize all the queries involving their relationship. In particular, we introduce a semantic-geometry disentangled query (SGDQ) design that enables the query features to attend separately to semantic information and geometric information relevant to the corresponding sub-tasks. A hybrid bipartite matching module is employed to well use the supervisions from all the sub-tasks during training. Qualitative and quantitative experimental results demonstrate that our method achieves state-of-the-art performance on the challenging ScanNet dataset. Code is available at https://github.com/SAITPublic/DOCTR.
☆ Re2LLM: Reflective Reinforcement Large Language Model for Session-based Recommendation
Large Language Models (LLMs) are emerging as promising approaches to enhance session-based recommendation (SBR), where both prompt-based and fine-tuning-based methods have been widely investigated to align LLMs with SBR. However, the former methods struggle with optimal prompts to elicit the correct reasoning of LLMs due to the lack of task-specific feedback, leading to unsatisfactory recommendations. Although the latter methods attempt to fine-tune LLMs with domain-specific knowledge, they face limitations such as high computational costs and reliance on open-source backbones. To address such issues, we propose a \underline{Re}flective \underline{Re}inforcement \underline{L}arge \underline{L}anguage \underline{M}odel (Re2LLM) for SBR, guiding LLMs to focus on specialized knowledge essential for more accurate recommendations effectively and efficiently. In particular, we first design the Reflective Exploration Module to effectively extract knowledge that is readily understandable and digestible by LLMs. To be specific, we direct LLMs to examine recommendation errors through self-reflection and construct a knowledge base (KB) comprising hints capable of rectifying these errors. To efficiently elicit the correct reasoning of LLMs, we further devise the Reinforcement Utilization Module to train a lightweight retrieval agent. It learns to select hints from the constructed KB based on the task-specific feedback, where the hints can serve as guidance to help correct LLMs reasoning for better recommendations. Extensive experiments on multiple real-world datasets demonstrate that our method consistently outperforms state-of-the-art methods.
comment: 11 pages, 4 figures
☆ An Experiment with the Use of ChatGPT for LCSH Subject Assignment on Electronic Theses and Dissertations
This study delves into the potential use of Large Language Models (LLMs) for generating Library of Congress Subject Headings (LCSH). The authors employed ChatGPT to generate subject headings for electronic theses and dissertations (ETDs) based on their titles and summaries. The results revealed that although some generated subject headings were valid, there were issues regarding specificity and exhaustiveness. The study showcases that LLMs can serve as a strategic response to the backlog of items awaiting cataloging in academic libraries, while also offering a cost-effective approach for promptly generating LCSH. Nonetheless, human catalogers remain essential for verifying and enhancing the validity, exhaustiveness, and specificity of LCSH generated by LLMs.
comment: 20 pages
☆ Refining Text-to-Image Generation: Towards Accurate Training-Free Glyph-Enhanced Image Generation
Over the past few years, Text-to-Image (T2I) generation approaches based on diffusion models have gained significant attention. However, vanilla diffusion models often suffer from spelling inaccuracies in the text displayed within the generated images. The capability to generate visual text is crucial, offering both academic interest and a wide range of practical applications. To produce accurate visual text images, state-of-the-art techniques adopt a glyph-controlled image generation approach, consisting of a text layout generator followed by an image generator that is conditioned on the generated text layout. Nevertheless, our study reveals that these models still face three primary challenges, prompting us to develop a testbed to facilitate future research. We introduce a benchmark, LenCom-Eval, specifically designed for testing models' capability in generating images with Lengthy and Complex visual text. Subsequently, we introduce a training-free framework to enhance the two-stage generation approaches. We examine the effectiveness of our approach on both LenCom-Eval and MARIO-Eval benchmarks and demonstrate notable improvements across a range of evaluation metrics, including CLIPScore, OCR precision, recall, F1 score, accuracy, and edit distance scores. For instance, our proposed framework improves the backbone model, TextDiffuser, by more than 23\% and 13.5\% in terms of OCR word F1 on LenCom-Eval and MARIO-Eval, respectively. Our work makes a unique contribution to the field by focusing on generating images with long and rare text sequences, a niche previously unexplored by existing literature
☆ An incremental MaxSAT-based model to learn balanced rules
The increasing advancements in the field of machine learning have led to the development of numerous applications that effectively address a wide range of problems with accurate predictions. However, in certain cases, accuracy alone may not be sufficient. Many real-world problems also demand explanations and interpretability behind the predictions. One of the most popular interpretable models that are classification rules. This work aims to propose an incremental model for learning interpretable and balanced rules based on MaxSAT, called IMLIB. This new model was based on two other approaches, one based on SAT and the other on MaxSAT. The one based on SAT limits the size of each generated rule, making it possible to balance them. We suggest that such a set of rules seem more natural to be understood compared to a mixture of large and small rules. The approach based on MaxSAT, called IMLI, presents a technique to increase performance that involves learning a set of rules by incrementally applying the model in a dataset. Finally, IMLIB and IMLI are compared using diverse databases. IMLIB obtained results comparable to IMLI in terms of accuracy, generating more balanced rules with smaller sizes.
comment: 16 pages, 5 tables, submitted to BRACIS 2023 (Brazilian Conference on Intelligent Systems), accepted version published in Intelligent Systems, LNCS, vol 14195
☆ How Reliable is Your Simulator? Analysis on the Limitations of Current LLM-based User Simulators for Conversational Recommendation
Conversational Recommender System (CRS) interacts with users through natural language to understand their preferences and provide personalized recommendations in real-time. CRS has demonstrated significant potential, prompting researchers to address the development of more realistic and reliable user simulators as a key focus. Recently, the capabilities of Large Language Models (LLMs) have attracted a lot of attention in various fields. Simultaneously, efforts are underway to construct user simulators based on LLMs. While these works showcase innovation, they also come with certain limitations that require attention. In this work, we aim to analyze the limitations of using LLMs in constructing user simulators for CRS, to guide future research. To achieve this goal, we conduct analytical validation on the notable work, iEvaLM. Through multiple experiments on two widely-used datasets in the field of conversational recommendation, we highlight several issues with the current evaluation methods for user simulators based on LLMs: (1) Data leakage, which occurs in conversational history and the user simulator's replies, results in inflated evaluation results. (2) The success of CRS recommendations depends more on the availability and quality of conversational history than on the responses from user simulators. (3) Controlling the output of the user simulator through a single prompt template proves challenging. To overcome these limitations, we propose SimpleUserSim, employing a straightforward strategy to guide the topic toward the target items. Our study validates the ability of CRS models to utilize the interaction information, significantly improving the recommendation results.
☆ Rethinking the Representation in Federated Unsupervised Learning with Non-IID Data CVPR 2024
Federated learning achieves effective performance in modeling decentralized data. In practice, client data are not well-labeled, which makes it potential for federated unsupervised learning (FUSL) with non-IID data. However, the performance of existing FUSL methods suffers from insufficient representations, i.e., (1) representation collapse entanglement among local and global models, and (2) inconsistent representation spaces among local models. The former indicates that representation collapse in local model will subsequently impact the global model and other local models. The latter means that clients model data representation with inconsistent parameters due to the deficiency of supervision signals. In this work, we propose FedU2 which enhances generating uniform and unified representation in FUSL with non-IID data. Specifically, FedU2 consists of flexible uniform regularizer (FUR) and efficient unified aggregator (EUA). FUR in each client avoids representation collapse via dispersing samples uniformly, and EUA in server promotes unified representation by constraining consistent client model updating. To extensively validate the performance of FedU2, we conduct both cross-device and cross-silo evaluation experiments on two benchmark datasets, i.e., CIFAR10 and CIFAR100.
comment: CVPR 2024
☆ RadioGAT: A Joint Model-based and Data-driven Framework for Multi-band Radiomap Reconstruction via Graph Attention Networks
Multi-band radiomap reconstruction (MB-RMR) is a key component in wireless communications for tasks such as spectrum management and network planning. However, traditional machine-learning-based MB-RMR methods, which rely heavily on simulated data or complete structured ground truth, face significant deployment challenges. These challenges stem from the differences between simulated and actual data, as well as the scarcity of real-world measurements. To address these challenges, our study presents RadioGAT, a novel framework based on Graph Attention Network (GAT) tailored for MB-RMR within a single area, eliminating the need for multi-region datasets. RadioGAT innovatively merges model-based spatial-spectral correlation encoding with data-driven radiomap generalization, thus minimizing the reliance on extensive data sources. The framework begins by transforming sparse multi-band data into a graph structure through an innovative encoding strategy that leverages radio propagation models to capture the spatial-spectral correlation inherent in the data. This graph-based representation not only simplifies data handling but also enables tailored label sampling during training, significantly enhancing the framework's adaptability for deployment. Subsequently, The GAT is employed to generalize the radiomap information across various frequency bands. Extensive experiments using raytracing datasets based on real-world environments have demonstrated RadioGAT's enhanced accuracy in supervised learning settings and its robustness in semi-supervised scenarios. These results underscore RadioGAT's effectiveness and practicality for MB-RMR in environments with limited data availability.
comment: submitted to IEEE journal for possible publication
☆ Skews in the Phenomenon Space Hinder Generalization in Text-to-Image Generation
The literature on text-to-image generation is plagued by issues of faithfully composing entities with relations. But there lacks a formal understanding of how entity-relation compositions can be effectively learned. Moreover, the underlying phenomenon space that meaningfully reflects the problem structure is not well-defined, leading to an arms race for larger quantities of data in the hope that generalization emerges out of large-scale pretraining. We hypothesize that the underlying phenomenological coverage has not been proportionally scaled up, leading to a skew of the presented phenomenon which harms generalization. We introduce statistical metrics that quantify both the linguistic and visual skew of a dataset for relational learning, and show that generalization failures of text-to-image generation are a direct result of incomplete or unbalanced phenomenological coverage. We first perform experiments in a synthetic domain and demonstrate that systematically controlled metrics are strongly predictive of generalization performance. Then we move to natural images and show that simple distribution perturbations in light of our theories boost generalization without enlarging the absolute data size. This work informs an important direction towards quality-enhancing the data diversity or balance orthogonal to scaling up the absolute size. Our discussions point out important open questions on 1) Evaluation of generated entity-relation compositions, and 2) Better models for reasoning with abstract relations.
☆ Concurrent Linguistic Error Detection (CLED) for Large Language Models
The wide adoption of Large language models (LLMs) makes their dependability a pressing concern. Detection of errors is the first step to mitigating their impact on a system and thus, efficient error detection for LLMs is an important issue. In many settings, the LLM is considered as a black box with no access to the internal nodes; this prevents the use of many error detection schemes that need access to the model's internal nodes. An interesting observation is that the output of LLMs in error-free operation should be valid and normal text. Therefore, when the text is not valid or differs significantly from normal text, it is likely that there is an error. Based on this observation we propose to perform Concurrent Linguistic Error Detection (CLED); this scheme extracts some linguistic features of the text generated by the LLM and feeds them to a concurrent classifier that detects errors. Since the proposed error detection mechanism only relies on the outputs of the model, then it can be used on LLMs in which there is no access to the internal nodes. The proposed CLED scheme has been evaluated on the T5 model when used for news summarization and on the OPUS-MT model when used for translation. In both cases, the same set of linguistic features has been used for error detection to illustrate the applicability of the proposed scheme beyond a specific case. The results show that CLED can detect most of the errors at a low overhead penalty. The use of the concurrent classifier also enables a trade-off between error detection effectiveness and its associated overhead, so providing flexibility to a designer.
comment: 11 pages, 6 figures, 30 references
☆ Dia-LLaMA: Towards Large Language Model-driven CT Report Generation
Medical report generation has achieved remarkable advancements yet has still been faced with several challenges. First, the inherent imbalance in the distribution of normal and abnormal cases may lead models to exhibit a biased focus on normal samples, resulting in unreliable diagnoses. Second, the frequent occurrence of common template sentences in the reports may overwhelm the critical abnormal information. Moreover, existing works focus on 2D chest X-rays, leaving CT report generation underexplored due to the high-dimensional nature of CT images and the limited availability of CT-report pairs. Recently, LLM has shown a great ability to generate reliable answers with appropriate prompts, which shed light on addressing the aforementioned challenges. In this paper, we propose Dia-LLaMA, a framework to adapt the LLaMA2-7B for CT report generation by incorporating diagnostic information as guidance prompts. Considering the high dimension of CT, we leverage a pre-trained ViT3D with perceiver to extract the visual information. To tailor the LLM for report generation and emphasize abnormality, we extract additional diagnostic information by referring to a disease prototype memory bank, which is updated during training to capture common disease representations. Furthermore, we introduce disease-aware attention to enable the model to adjust attention for different diseases. Experiments on the chest CT dataset demonstrated that our proposed method outperformed previous methods and achieved state-of-the-art on both clinical efficacy performance and natural language generation metrics. The code will be made publically available.
comment: 10 pages
☆ Learning Action-based Representations Using Invariance
Robust reinforcement learning agents using high-dimensional observations must be able to identify relevant state features amidst many exogeneous distractors. A representation that captures controllability identifies these state elements by determining what affects agent control. While methods such as inverse dynamics and mutual information capture controllability for a limited number of timesteps, capturing long-horizon elements remains a challenging problem. Myopic controllability can capture the moment right before an agent crashes into a wall, but not the control-relevance of the wall while the agent is still some distance away. To address this we introduce action-bisimulation encoding, a method inspired by the bisimulation invariance pseudometric, that extends single-step controllability with a recursive invariance constraint. By doing this, action-bisimulation learns a multi-step controllability metric that smoothly discounts distant state features that are relevant for control. We demonstrate that action-bisimulation pretraining on reward-free, uniformly random data improves sample efficiency in several environments, including a photorealistic 3D simulation domain, Habitat. Additionally, we provide theoretical analysis and qualitative results demonstrating the information captured by action-bisimulation.
☆ ChatDBG: An AI-Powered Debugging Assistant
This paper presents ChatDBG, the first AI-powered debugging assistant. ChatDBG integrates large language models (LLMs) to significantly enhance the capabilities and user-friendliness of conventional debuggers. ChatDBG lets programmers engage in a collaborative dialogue with the debugger, allowing them to pose complex questions about program state, perform root cause analysis for crashes or assertion failures, and explore open-ended queries like "why is x null?". To handle these queries, ChatDBG grants the LLM autonomy to take the wheel and drive debugging by issuing commands to navigate through stacks and inspect program state; it then reports its findings and yields back control to the programmer. Our ChatDBG prototype integrates with standard debuggers including LLDB, GDB, and WinDBG for native code and Pdb for Python. Our evaluation across a diverse set of code, including C/C++ code with known bugs and a suite of Python code including standalone scripts and Jupyter notebooks, demonstrates that ChatDBG can successfully analyze root causes, explain bugs, and generate accurate fixes for a wide range of real-world errors. For the Python programs, a single query led to an actionable bug fix 67% of the time; one additional follow-up query increased the success rate to 85%. ChatDBG has seen rapid uptake; it has already been downloaded nearly 30,000 times.
comment: 11 pages
☆ ChatGPT Incorrectness Detection in Software Reviews
We conducted a survey of 135 software engineering (SE) practitioners to understand how they use Generative AI-based chatbots like ChatGPT for SE tasks. We find that they want to use ChatGPT for SE tasks like software library selection but often worry about the truthfulness of ChatGPT responses. We developed a suite of techniques and a tool called CID (ChatGPT Incorrectness Detector) to automatically test and detect the incorrectness in ChatGPT responses. CID is based on the iterative prompting to ChatGPT by asking it contextually similar but textually divergent questions (using an approach that utilizes metamorphic relationships in texts). The underlying principle in CID is that for a given question, a response that is different from other responses (across multiple incarnations of the question) is likely an incorrect response. In a benchmark study of library selection, we show that CID can detect incorrect responses from ChatGPT with an F1-score of 0.74 - 0.75.
☆ Enhanced Facet Generation with LLM Editing LREC
In information retrieval, facet identification of a user query is an important task. If a search service can recognize the facets of a user's query, it has the potential to offer users a much broader range of search results. Previous studies can enhance facet prediction by leveraging retrieved documents and related queries obtained through a search engine. However, there are challenges in extending it to other applications when a search engine operates as part of the model. First, search engines are constantly updated. Therefore, additional information may change during training and test, which may reduce performance. The second challenge is that public search engines cannot search for internal documents. Therefore, a separate search system needs to be built to incorporate documents from private domains within the company. We propose two strategies that focus on a framework that can predict facets by taking only queries as input without a search engine. The first strategy is multi-task learning to predict SERP. By leveraging SERP as a target instead of a source, the proposed model deeply understands queries without relying on external modules. The second strategy is to enhance the facets by combining Large Language Model (LLM) and the small model. Overall performance improves when small model and LLM are combined rather than facet generation individually.
comment: Accepted at LREC-COLING 2024
☆ Impact of Video Compression Artifacts on Fisheye Camera Visual Perception Tasks
Autonomous driving systems require extensive data collection schemes to cover the diverse scenarios needed for building a robust and safe system. The data volumes are in the order of Exabytes and have to be stored for a long period of time (i.e., more than 10 years of the vehicle's life cycle). Lossless compression doesn't provide sufficient compression ratios, hence, lossy video compression has been explored. It is essential to prove that lossy video compression artifacts do not impact the performance of the perception algorithms. However, there is limited work in this area to provide a solid conclusion. In particular, there is no such work for fisheye cameras, which have high radial distortion and where compression may have higher artifacts. Fisheye cameras are commonly used in automotive systems for 3D object detection task. In this work, we provide the first analysis of the impact of standard video compression codecs on wide FOV fisheye camera images. We demonstrate that the achievable compression with negligible impact depends on the dataset and temporal prediction of the video codec. We propose a radial distortion-aware zonal metric to evaluate the performance of artifacts in fisheye images. In addition, we present a novel method for estimating affine mode parameters of the latest VVC codec, and suggest some areas for improvement in video codecs for the application to fisheye imagery.
☆ Graphs Generalization under Distribution Shifts
Traditional machine learning methods heavily rely on the independent and identically distribution assumption, which imposes limitations when the test distribution deviates from the training distribution. To address this crucial issue, out-of-distribution (OOD) generalization, which aims to achieve satisfactory generalization performance when faced with unknown distribution shifts, has made a significant process. However, the OOD method for graph-structured data currently lacks clarity and remains relatively unexplored due to two primary challenges. Firstly, distribution shifts on graphs often occur simultaneously on node attributes and graph topology. Secondly, capturing invariant information amidst diverse distribution shifts proves to be a formidable challenge. To overcome these obstacles, in this paper, we introduce a novel framework, namely Graph Learning Invariant Domain genERation (GLIDER). The goal is to (1) diversify variations across domains by modeling the potential seen or unseen variations of attribute distribution and topological structure and (2) minimize the discrepancy of the variation in a representation space where the target is to predict semantic labels. Extensive experiment results indicate that our model outperforms baseline methods on node-level OOD generalization across domains in distribution shift on node features and topological structures simultaneously.
♻ ☆ Word4Per: Zero-shot Composed Person Retrieval
Searching for specific person has great social benefits and security value, and it often involves a combination of visual and textual information. Conventional person retrieval methods, whether image-based or text-based, usually fall short in effectively harnessing both types of information, leading to the loss of accuracy. In this paper, a whole new task called Composed Person Retrieval (CPR) is proposed to jointly utilize both image and text information for target person retrieval. However, the supervised CPR requires very costly manual annotation dataset, while there are currently no available resources. To mitigate this issue, we firstly introduce the Zero-shot Composed Person Retrieval (ZS-CPR), which leverages existing domain-related data to resolve the CPR problem without expensive annotations. Secondly, to learn ZS-CPR model, we propose a two-stage learning framework, Word4Per, where a lightweight Textual Inversion Network (TINet) and a text-based person retrieval model based on fine-tuned Contrastive Language-Image Pre-training (CLIP) network are learned without utilizing any CPR data. Thirdly, a finely annotated Image-Text Composed Person Retrieval (ITCPR) dataset is built as the benchmark to assess the performance of the proposed Word4Per framework. Extensive experiments under both Rank-1 and mAP demonstrate the effectiveness of Word4Per for the ZS-CPR task, surpassing the comparative methods by over 10\%. The code and ITCPR dataset will be publicly available at https://github.com/Delong-liu-bupt/Word4Per.
♻ ☆ Design-Space Exploration of SNN Models using Application-Specific Multi-Core Architectures
With the motivation and the difficulties that currently exist in comprehending and utilizing the promising features of SNNs, we proposed a novel run-time multi-core architecture-based simulator called "RAVSim" (Runtime Analysis and Visualization Simulator), a cutting-edge SNN simulator, developed using LabVIEW and it is publicly available on their website as an official module. RAVSim is a runtime virtual simulation environment tool that enables the user to interact with the model, observe its behavior of output concentration, and modify the set of parametric values at any time while the simulation is in execution. Recently some popular tools have been presented, but we believe that none of the tools allow users to interact with the model simulation in run time.
comment: Abstract Presentation in 2023 Neuro-Inspired Computing Elements (NICE) Conference
♻ ☆ LongHeads: Multi-Head Attention is Secretly a Long Context Processor
Large language models (LLMs) have achieved impressive performance in numerous domains but often struggle to process lengthy inputs effectively and efficiently due to limited length generalization and attention's quadratic computational demands. Many sought to mitigate this by restricting the attention window within the pre-trained length. However, these methods introduce new issues such as ignoring the middle context and requiring additional training. To address these problems, we propose LongHeads, a training-free framework that enhances LLM's long context ability by unlocking multi-head attention's untapped potential. Instead of allowing each head to attend to the full sentence, which struggles with generalizing to longer sequences due to out-of-distribution (OOD) issues, we allow each head to process in-distribution length by selecting and attending to important context chunks. To this end, we propose a chunk selection strategy that relies on the inherent correlation between the query and the key representations, efficiently distributing context chunks to different heads. In this way, each head ensures it can effectively process attended tokens within the trained length, while different heads in different layers can collectively process longer contexts. LongHeads works efficiently in linear time, fits seamlessly with many LLMs that use relative positional encoding. LongHeads achieves 100% accuracy at the 128k length on passkey retrieval task, verifying LongHeads's efficacy in extending the usable context window for existing models. We release our code at https://github.com/LuLuLuyi/LongHeads .
♻ ☆ SVGDreamer: Text Guided SVG Generation with Diffusion Model CVPR 2024
Recently, text-guided scalable vector graphics (SVGs) synthesis has shown promise in domains such as iconography and sketch. However, existing text-to-SVG generation methods lack editability and struggle with visual quality and result diversity. To address these limitations, we propose a novel text-guided vector graphics synthesis method called SVGDreamer. SVGDreamer incorporates a semantic-driven image vectorization (SIVE) process that enables the decomposition of synthesis into foreground objects and background, thereby enhancing editability. Specifically, the SIVE process introduce attention-based primitive control and an attention-mask loss function for effective control and manipulation of individual elements. Additionally, we propose a Vectorized Particle-based Score Distillation (VPSD) approach to tackle the challenges of shape over-smoothing, color over-saturation, limited diversity in results, and slow convergence in existing text-to-SVG generation methods. VPSD models SVGs as distributions of control points and colors to counteract over-smoothing and over-saturation. Furthermore, VPSD leverages a reward model to reweight vector particles, which improves aesthetic appeal and accelerates convergence. Extensive experiments have been conducted to validate the effectiveness of SVGDreamer, demonstrating its superiority over baseline methods in terms of editability, visual quality, and diversity. The code and demo of SVGDreamer can be found at https://ximinng.github.io/SVGDreamer-project/
comment: Accepted by CVPR 2024. project link: https://ximinng.github.io/SVGDreamer-project/
♻ ☆ The Implications of Decentralization in Blockchained Federated Learning: Evaluating the Impact of Model Staleness and Inconsistencies
Blockchain promises to enhance distributed machine learning (ML) approaches such as federated learning (FL) by providing further decentralization, security, immutability, and trust, which are key properties for enabling collaborative intelligence in next-generation applications. Nonetheless, the intrinsic decentralized operation of peer-to-peer (P2P) blockchain nodes leads to an uncharted setting for FL, whereby the concepts of FL round and global model become meaningless, as devices' synchronization is lost without the figure of a central orchestrating server. In this paper, we study the practical implications of outsourcing the orchestration of FL to a democratic setting such as in a blockchain. In particular, we focus on the effects that model staleness and inconsistencies, endorsed by blockchains' modus operandi, have on the training procedure held by FL devices asynchronously. Using simulation, we evaluate the blockchained FL operation by applying two different ML models (ranging from low to high complexity) on the well-known MNIST and CIFAR-10 datasets, respectively, and focus on the accuracy and timeliness of the solutions. Our results show the high impact of model inconsistencies on the accuracy of the models (up to a ~35% decrease in prediction accuracy), which underscores the importance of properly designing blockchain systems based on the characteristics of the underlying FL application.
♻ ☆ Sandra -- A Neuro-Symbolic Reasoner Based On Descriptions And Situations
This paper presents sandra, a neuro-symbolic reasoner combining vectorial representations with deductive reasoning. Sandra builds a vector space constrained by an ontology and performs reasoning over it. The geometric nature of the reasoner allows its combination with neural networks, bridging the gap with symbolic knowledge representations. Sandra is based on the Description and Situation (DnS) ontology design pattern, a formalization of frame semantics. Given a set of facts (a situation) it allows to infer all possible perspectives (descriptions) that can provide a plausible interpretation for it, even in presence of incomplete information. We prove that our method is correct with respect to the DnS model. We experiment with two different tasks and their standard benchmarks, demonstrating that, without increasing complexity, sandra (i) outperforms all the baselines (ii) provides interpretability in the classification process, and (iii) allows control over the vector space, which is designed a priori.
♻ ☆ BatteryML:An Open-source platform for Machine Learning on Battery Degradation
Battery degradation remains a pivotal concern in the energy storage domain, with machine learning emerging as a potent tool to drive forward insights and solutions. However, this intersection of electrochemical science and machine learning poses complex challenges. Machine learning experts often grapple with the intricacies of battery science, while battery researchers face hurdles in adapting intricate models tailored to specific datasets. Beyond this, a cohesive standard for battery degradation modeling, inclusive of data formats and evaluative benchmarks, is conspicuously absent. Recognizing these impediments, we present BatteryML - a one-step, all-encompass, and open-source platform designed to unify data preprocessing, feature extraction, and the implementation of both traditional and state-of-the-art models. This streamlined approach promises to enhance the practicality and efficiency of research applications. BatteryML seeks to fill this void, fostering an environment where experts from diverse specializations can collaboratively contribute, thus elevating the collective understanding and advancement of battery research.The code for our project is publicly available on GitHub at https://github.com/microsoft/BatteryML.
♻ ☆ Causal Question Answering with Reinforcement Learning WWW 2024
Causal questions inquire about causal relationships between different events or phenomena. They are important for a variety of use cases, including virtual assistants and search engines. However, many current approaches to causal question answering cannot provide explanations or evidence for their answers. Hence, in this paper, we aim to answer causal questions with a causality graph, a large-scale dataset of causal relations between noun phrases along with the relations' provenance data. Inspired by recent, successful applications of reinforcement learning to knowledge graph tasks, such as link prediction and fact-checking, we explore the application of reinforcement learning on a causality graph for causal question answering. We introduce an Actor-Critic-based agent which learns to search through the graph to answer causal questions. We bootstrap the agent with a supervised learning procedure to deal with large action spaces and sparse rewards. Our evaluation shows that the agent successfully prunes the search space to answer binary causal questions by visiting less than 30 nodes per question compared to over 3,000 nodes by a naive breadth-first search. Our ablation study indicates that our supervised learning strategy provides a strong foundation upon which our reinforcement learning agent improves. The paths returned by our agent explain the mechanisms by which a cause produces an effect. Moreover, for each edge on a path, our causality graph provides its original source allowing for easy verification of paths.
comment: Accepted at WWW 2024
♻ ☆ Align-to-Distill: Trainable Attention Alignment for Knowledge Distillation in Neural Machine Translation LREC
The advent of scalable deep models and large datasets has improved the performance of Neural Machine Translation. Knowledge Distillation (KD) enhances efficiency by transferring knowledge from a teacher model to a more compact student model. However, KD approaches to Transformer architecture often rely on heuristics, particularly when deciding which teacher layers to distill from. In this paper, we introduce the 'Align-to-Distill' (A2D) strategy, designed to address the feature mapping problem by adaptively aligning student attention heads with their teacher counterparts during training. The Attention Alignment Module in A2D performs a dense head-by-head comparison between student and teacher attention heads across layers, turning the combinatorial mapping heuristics into a learning problem. Our experiments show the efficacy of A2D, demonstrating gains of up to +3.61 and +0.63 BLEU points for WMT-2022 De->Dsb and WMT-2014 En->De, respectively, compared to Transformer baselines.
comment: Accepted to LREC-COLING 2024
♻ ☆ Exploring the Adversarial Capabilities of Large Language Models
The proliferation of large language models (LLMs) has sparked widespread and general interest due to their strong language generation capabilities, offering great potential for both industry and research. While previous research delved into the security and privacy issues of LLMs, the extent to which these models can exhibit adversarial behavior remains largely unexplored. Addressing this gap, we investigate whether common publicly available LLMs have inherent capabilities to perturb text samples to fool safety measures, so-called adversarial examples resp.~attacks. More specifically, we investigate whether LLMs are inherently able to craft adversarial examples out of benign samples to fool existing safe rails. Our experiments, which focus on hate speech detection, reveal that LLMs succeed in finding adversarial perturbations, effectively undermining hate speech detection systems. Our findings carry significant implications for (semi-)autonomous systems relying on LLMs, highlighting potential challenges in their interaction with existing systems and safety measures.
♻ ☆ HealthFC: Verifying Health Claims with Evidence-Based Medical Fact-Checking LREC
In the digital age, seeking health advice on the Internet has become a common practice. At the same time, determining the trustworthiness of online medical content is increasingly challenging. Fact-checking has emerged as an approach to assess the veracity of factual claims using evidence from credible knowledge sources. To help advance automated Natural Language Processing (NLP) solutions for this task, in this paper we introduce a novel dataset HealthFC. It consists of 750 health-related claims in German and English, labeled for veracity by medical experts and backed with evidence from systematic reviews and clinical trials. We provide an analysis of the dataset, highlighting its characteristics and challenges. The dataset can be used for NLP tasks related to automated fact-checking, such as evidence retrieval, claim verification, or explanation generation. For testing purposes, we provide baseline systems based on different approaches, examine their performance, and discuss the findings. We show that the dataset is a challenging test bed with a high potential for future use.
comment: Accepted to LREC-COLING 2024
♻ ☆ With Greater Text Comes Greater Necessity: Inference-Time Training Helps Long Text Generation
Long text generation, such as novel writing and discourse-level translation with extremely long contexts, presents significant challenges to current language models. Existing methods mainly focus on extending the model's context window through strategies like length extrapolation. However, these approaches demand substantial hardware resources during the training and/or inference phases. Our proposed method, Temp-Lora, introduces an alternative concept. Instead of relying on the KV cache to store all context information, we embeds this information directly into a temporary Lora module. In the process of long text generation, this module is progressively trained with text generated previously. This approach not only efficiently preserves contextual knowledge but also prevents any permanent alteration to the model's parameters given that the module is discarded post-generation. Extensive experiments on the PG19 language modeling benchmark and the GuoFeng discourse-level translation benchmark validate the effectiveness of Temp-Lora. Our results show that: 1) Temp-Lora substantially enhances generation quality for long text, as indicated by a 13.2% decrease in perplexity (PPL) on a subset of PG19, and a 29.3% decrease in PPL along with a 113.2% increase in BLEU score on a subset of GuoFeng, 2) Temp-Lora is compatible with and enhances most existing long text generation methods, and 3) Temp-Lora can greatly reduce computational costs by shortening the context window. For example, we can ensure a moderate improvement in generation quality (a decrease of 3.8% in PPL) while enabling a 51.5% memory usage reduction and a 60.0% decrease in latency for inference.
♻ ☆ Multi-agent reinforcement learning using echo-state network and its application to pedestrian dynamics
In recent years, simulations of pedestrians using the multi-agent reinforcement learning (MARL) have been studied. This study considered the roads on a grid-world environment, and implemented pedestrians as MARL agents using an echo-state network and the least squares policy iteration method. Under this environment, the ability of these agents to learn to move forward by avoiding other agents was investigated. Specifically, we considered two types of tasks: the choice between a narrow direct route and a broad detour, and the bidirectional pedestrian flow in a corridor. The simulations results indicated that the learning was successful when the density of the agents was not that high.
comment: 26 pages, 17 figures
♻ ☆ Dial-MAE: ConTextual Masked Auto-Encoder for Retrieval-based Dialogue Systems NAACL 2024
Dialogue response selection aims to select an appropriate response from several candidates based on a given user and system utterance history. Most existing works primarily focus on post-training and fine-tuning tailored for cross-encoders. However, there are no post-training methods tailored for dense encoders in dialogue response selection. We argue that when the current language model, based on dense dialogue systems (such as BERT), is employed as a dense encoder, it separately encodes dialogue context and response, leading to a struggle to achieve the alignment of both representations. Thus, we propose Dial-MAE (Dialogue Contextual Masking Auto-Encoder), a straightforward yet effective post-training technique tailored for dense encoders in dialogue response selection. Dial-MAE uses an asymmetric encoder-decoder architecture to compress the dialogue semantics into dense vectors, which achieves better alignment between the features of the dialogue context and response. Our experiments have demonstrated that Dial-MAE is highly effective, achieving state-of-the-art performance on two commonly evaluated benchmarks.
comment: This paper has been accepted by NAACL 2024
♻ ☆ Preference as Reward, Maximum Preference Optimization with Importance Sampling
Preference learning is a key technology for aligning language models with human values. Reinforcement Learning from Human Feedback (RLHF) is a model-based algorithm to optimize preference learning, which first fits a reward model for preference scores and then optimizes the generating policy with an on-policy PPO algorithm to maximize the reward. The processing of RLHF is complex, time-consuming, and unstable. The Direct Preference Optimization (DPO) algorithm uses an off-policy algorithm to directly optimize the generating policy and eliminates the need for a reward model. DPO is more data-efficient and stable. However, DPO has a drawback of overfitting to the preference data and ignoring the KL-regularization term when the preference is deterministic. Identity mapping Preference Optimization(IPO) uses a root-finding MSE loss to incorporate KL-regularization. However, both DPO and IPO fail to properly address the KL-regularization term because the support of the preference distribution is not equal to the reference distribution. In this paper, we propose a simple and intuitive off-policy preference optimization algorithm from an importance sampling view, which we call Maximum Preference Optimization (MPO). MPO incorporates the off-policy KL-regularization term, making regularization truly effective. MPO achieves the best of both worlds by combining the objectives of RLHF and IPO while being an off-policy algorithm. Furthermore, MPO eliminates the need for a reward model and reference policy, simplifying the learning process and reducing memory usage.
♻ ☆ A Survey of Confidence Estimation and Calibration in Large Language Models
Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks in various domains. Despite their impressive performance, they can be unreliable due to factual errors in their generations. Assessing their confidence and calibrating them across different tasks can help mitigate risks and enable LLMs to produce better generations. There has been a lot of recent research aiming to address this, but there has been no comprehensive overview to organize it and outline the main lessons learned. The present survey aims to bridge this gap. In particular, we outline the challenges and we summarize recent technical advancements for LLM confidence estimation and calibration. We further discuss their applications and suggest promising directions for future work.
comment: 16 pages, 1 page, 1 table
♻ ☆ A Knowledge Engineering Primer
The aim of this primer is to introduce the subject of knowledge engineering in a concise but synthetic way to develop the reader's intuition about the area.
♻ ☆ Exploring ChatGPT and its Impact on Society
Artificial intelligence has been around for a while, but suddenly it has received more attention than ever before. Thanks to innovations from companies like Google, Microsoft, Meta, and other major brands in technology. OpenAI, though, has triggered the button with its ground-breaking invention ChatGPT. ChatGPT is a Large Language Model (LLM) based on Transformer architecture that has the ability to generate human-like responses in a conversational context. It uses deep learning algorithms to generate natural language responses to input text. Its large number of parameters, contextual generation, and open-domain training make it a versatile and effective tool for a wide range of applications, from chatbots to customer service to language translation. It has the potential to revolutionize various industries and transform the way we interact with technology. However, the use of ChatGPT has also raised several concerns, including ethical, social, and employment challenges, which must be carefully considered to ensure the responsible use of this technology. The article provides an overview of ChatGPT, delving into its architecture and training process. It highlights the potential impacts of ChatGPT on the society. In this paper, we suggest some approaches involving technology, regulation, education, and ethics in an effort to maximize ChatGPT's benefits while minimizing its negative impacts. This study is expected to contribute to a greater understanding of ChatGPT and aid in predicting the potential changes it may bring about.
comment: 13 Pages
♻ ☆ PIA: Your Personalized Image Animator via Plug-and-Play Modules in Text-to-Image Models
Recent advancements in personalized text-to-image (T2I) models have revolutionized content creation, empowering non-experts to generate stunning images with unique styles. While promising, adding realistic motions into these personalized images by text poses significant challenges in preserving distinct styles, high-fidelity details, and achieving motion controllability by text. In this paper, we present PIA, a Personalized Image Animator that excels in aligning with condition images, achieving motion controllability by text, and the compatibility with various personalized T2I models without specific tuning. To achieve these goals, PIA builds upon a base T2I model with well-trained temporal alignment layers, allowing for the seamless transformation of any personalized T2I model into an image animation model. A key component of PIA is the introduction of the condition module, which utilizes the condition frame and inter-frame affinity as input to transfer appearance information guided by the affinity hint for individual frame synthesis in the latent space. This design mitigates the challenges of appearance-related image alignment within and allows for a stronger focus on aligning with motion-related guidance.
comment: Project page: https://pi-animator.github.io/
♻ ☆ I-PHYRE: Interactive Physical Reasoning ICLR 2024
Current evaluation protocols predominantly assess physical reasoning in stationary scenes, creating a gap in evaluating agents' abilities to interact with dynamic events. While contemporary methods allow agents to modify initial scene configurations and observe consequences, they lack the capability to interact with events in real time. To address this, we introduce I-PHYRE, a framework that challenges agents to simultaneously exhibit intuitive physical reasoning, multi-step planning, and in-situ intervention. Here, intuitive physical reasoning refers to a quick, approximate understanding of physics to address complex problems; multi-step denotes the need for extensive sequence planning in I-PHYRE, considering each intervention can significantly alter subsequent choices; and in-situ implies the necessity for timely object manipulation within a scene, where minor timing deviations can result in task failure. We formulate four game splits to scrutinize agents' learning and generalization of essential principles of interactive physical reasoning, fostering learning through interaction with representative scenarios. Our exploration involves three planning strategies and examines several supervised and reinforcement agents' zero-shot generalization proficiency on I-PHYRE. The outcomes highlight a notable gap between existing learning algorithms and human performance, emphasizing the imperative for more research in enhancing agents with interactive physical reasoning capabilities. The environment and baselines will be made publicly available.
comment: 21 pages, ICLR 2024
♻ ☆ Developing and Deploying Industry Standards for Artificial Intelligence in Education (AIED): Challenges, Strategies, and Future Directions
The adoption of Artificial Intelligence in Education (AIED) holds the promise of revolutionizing educational practices by offering personalized learning experiences, automating administrative and pedagogical tasks, and reducing the cost of content creation. However, the lack of standardized practices in the development and deployment of AIED solutions has led to fragmented ecosystems, which presents challenges in interoperability, scalability, and ethical governance. This article aims to address the critical need to develop and implement industry standards in AIED, offering a comprehensive analysis of the current landscape, challenges, and strategic approaches to overcome these obstacles. We begin by examining the various applications of AIED in various educational settings and identify key areas lacking in standardization, including system interoperability, ontology mapping, data integration, evaluation, and ethical governance. Then, we propose a multi-tiered framework for establishing robust industry standards for AIED. In addition, we discuss methodologies for the iterative development and deployment of standards, incorporating feedback loops from real-world applications to refine and adapt standards over time. The paper also highlights the role of emerging technologies and pedagogical theories in shaping future standards for AIED. Finally, we outline a strategic roadmap for stakeholders to implement these standards, fostering a cohesive and ethical AIED ecosystem. By establishing comprehensive industry standards, such as those by IEEE Artificial Intelligence Standards Committee (AISC) and International Organization for Standardization (ISO), we can accelerate and scale AIED solutions to improve educational outcomes, ensuring that technological advances align with the principles of inclusivity, fairness, and educational excellence.
comment: 12 pages
♻ ☆ CiPR: An Efficient Framework with Cross-instance Positive Relations for Generalized Category Discovery
We tackle the issue of generalized category discovery (GCD). GCD considers the open-world problem of automatically clustering a partially labelled dataset, in which the unlabelled data may contain instances from both novel categories and labelled classes. In this paper, we address the GCD problem with an unknown category number for the unlabelled data. We propose a framework, named CiPR, to bootstrap the representation by exploiting Cross-instance Positive Relations in the partially labelled data for contrastive learning, which have been neglected in existing methods. To obtain reliable cross-instance relations to facilitate representation learning, we introduce a semi-supervised hierarchical clustering algorithm, named selective neighbor clustering (SNC), which can produce a clustering hierarchy directly from the connected components of a graph constructed from selective neighbors. We further present a method to estimate the unknown class number using SNC with a joint reference score that considers clustering indexes of both labelled and unlabelled data, and extend SNC to allow label assignment for the unlabelled instances with a given class number. We thoroughly evaluate our framework on public generic image recognition datasets and challenging fine-grained datasets, and establish a new state-of-the-art. Code: https://github.com/haoosz/CiPR
comment: Accepted to TMLR. Code: https://github.com/haoosz/CiPR
♻ ☆ HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data CVPR 2024
Multi-modal Large Language Models (MLLMs) tuned on machine-generated instruction-following data have demonstrated remarkable performance in various multi-modal understanding and generation tasks. However, the hallucinations inherent in machine-generated data, which could lead to hallucinatory outputs in MLLMs, remain under-explored. This work aims to investigate various hallucinations (i.e., object, relation, attribute hallucinations) and mitigate those hallucinatory toxicities in large-scale machine-generated visual instruction datasets. Drawing on the human ability to identify factual errors, we present a novel hallucination detection and elimination framework, HalluciDoctor, based on the cross-checking paradigm. We use our framework to identify and eliminate hallucinations in the training data automatically. Interestingly, HalluciDoctor also indicates that spurious correlations arising from long-tail object co-occurrences contribute to hallucinations. Based on that, we execute counterfactual visual instruction expansion to balance data distribution, thereby enhancing MLLMs' resistance to hallucinations. Comprehensive experiments on hallucination evaluation benchmarks show that our method successfully mitigates 44.6% hallucinations relatively and maintains competitive performance compared to LLaVA. The data and code for this paper are publicly available. \url{https://github.com/Yuqifan1117/HalluciDoctor}.
comment: Accepted by CVPR 2024
♻ ☆ Solving Data-centric Tasks using Large Language Models NAACL 2024
Large language models (LLMs) are rapidly replacing help forums like StackOverflow, and are especially helpful for non-professional programmers and end users. These users are often interested in data-centric tasks, such as spreadsheet manipulation and data wrangling, which are hard to solve if the intent is only communicated using a natural-language description, without including the data. But how do we decide how much data and which data to include in the prompt? This paper makes two contributions towards answering this question. First, we create a dataset of real-world NL-to-code tasks manipulating tabular data, mined from StackOverflow posts. Second, we introduce a cluster-then-select prompting technique, which adds the most representative rows from the input data to the LLM prompt. Our experiments show that LLM performance is indeed sensitive to the amount of data passed in the prompt, and that for tasks with a lot of syntactic variation in the input table, our cluster-then-select technique outperforms a random selection baseline.
comment: Paper accepted to NAACL 2024 (Findings)
♻ ☆ A Survey on Large Language Model based Autonomous Agents
Autonomous agents have long been a prominent research focus in both academic and industry communities. Previous research in this field often focuses on training agents with limited knowledge within isolated environments, which diverges significantly from human learning processes, and thus makes the agents hard to achieve human-like decisions. Recently, through the acquisition of vast amounts of web knowledge, large language models (LLMs) have demonstrated remarkable potential in achieving human-level intelligence. This has sparked an upsurge in studies investigating LLM-based autonomous agents. In this paper, we present a comprehensive survey of these studies, delivering a systematic review of the field of LLM-based autonomous agents from a holistic perspective. More specifically, we first discuss the construction of LLM-based autonomous agents, for which we propose a unified framework that encompasses a majority of the previous work. Then, we present a comprehensive overview of the diverse applications of LLM-based autonomous agents in the fields of social science, natural science, and engineering. Finally, we delve into the evaluation strategies commonly used for LLM-based autonomous agents. Based on the previous studies, we also present several challenges and future directions in this field. To keep track of this field and continuously update our survey, we maintain a repository of relevant references at https://github.com/Paitesanshi/LLM-Agent-Survey.
comment: 35 pages, 5 figures, 3 tables, has been accepted by frontiers of computer science (FCS), doi={10.1007/s11704-024-40231-1}
♻ ☆ Don't Judge by the Look: Towards Motion Coherent Video Representation ICLR2024
Current training pipelines in object recognition neglect Hue Jittering when doing data augmentation as it not only brings appearance changes that are detrimental to classification, but also the implementation is inefficient in practice. In this study, we investigate the effect of hue variance in the context of video understanding and find this variance to be beneficial since static appearances are less important in videos that contain motion information. Based on this observation, we propose a data augmentation method for video understanding, named Motion Coherent Augmentation (MCA), that introduces appearance variation in videos and implicitly encourages the model to prioritize motion patterns, rather than static appearances. Concretely, we propose an operation SwapMix to efficiently modify the appearance of video samples, and introduce Variation Alignment (VA) to resolve the distribution shift caused by SwapMix, enforcing the model to learn appearance invariant representations. Comprehensive empirical evaluation across various architectures and different datasets solidly validates the effectiveness and generalization ability of MCA, and the application of VA in other augmentation methods. Code is available at https://github.com/BeSpontaneous/MCA-pytorch.
comment: Accepted by ICLR2024
♻ ☆ EVOTER: Evolution of Transparent Explainable Rule-sets
Most AI systems are black boxes generating reasonable outputs for given inputs. Some domains, however, have explainability and trustworthiness requirements that cannot be directly met by these approaches. Various methods have therefore been developed to interpret black-box models after training. This paper advocates an alternative approach where the models are transparent and explainable to begin with. This approach, EVOTER, evolves rule-sets based on simple logical expressions. The approach is evaluated in several prediction/classification and prescription/policy search domains with and without a surrogate. It is shown to discover meaningful rule sets that perform similarly to black-box models. The rules can provide insight into the domain, and make biases hidden in the data explicit. It may also be possible to edit them directly to remove biases and add constraints. EVOTER thus forms a promising foundation for building trustworthy AI systems for real-world applications in the future.
♻ ☆ Cartoon Hallucinations Detection: Pose-aware In Context Visual Learning
Large-scale Text-to-Image (TTI) models have become a common approach for generating training data in various generative fields. However, visual hallucinations, which contain perceptually critical defects, remain a concern, especially in non-photorealistic styles like cartoon characters. We propose a novel visual hallucination detection system for cartoon character images generated by TTI models. Our approach leverages pose-aware in-context visual learning (PA-ICVL) with Vision-Language Models (VLMs), utilizing both RGB images and pose information. By incorporating pose guidance from a fine-tuned pose estimator, we enable VLMs to make more accurate decisions. Experimental results demonstrate significant improvements in identifying visual hallucinations compared to baseline methods relying solely on RGB images. This research advances TTI models by mitigating visual hallucinations, expanding their potential in non-photorealistic domains.
comment: 11 pages, 12 figures, 1 table, Project page: https://gh-bumsookim.github.io/Cartoon-Hallucinations-Detection/
♻ ☆ The AI Assessment Scale (AIAS) in action: A pilot implementation of GenAI supported assessment
The rapid adoption of Generative Artificial Intelligence (GenAI) technologies in higher education has raised concerns about academic integrity, assessment practices, and student learning. Banning or blocking GenAI tools has proven ineffective, and punitive approaches ignore the potential benefits of these technologies. This paper presents the findings of a pilot study conducted at British University Vietnam (BUV) exploring the implementation of the Artificial Intelligence Assessment Scale (AIAS), a flexible framework for incorporating GenAI into educational assessments. The AIAS consists of five levels, ranging from 'No AI' to 'Full AI', enabling educators to design assessments that focus on areas requiring human input and critical thinking. Following the implementation of the AIAS, the pilot study results indicate a significant reduction in academic misconduct cases related to GenAI, a 5.9% increase in student attainment across the university, and a 33.3% increase in module passing rates. The AIAS facilitated a shift in pedagogical practices, with faculty members incorporating GenAI tools into their modules and students producing innovative multimodal submissions. The findings suggest that the AIAS can support the effective integration of GenAI in HE, promoting academic integrity while leveraging the technology's potential to enhance learning experiences.
Optimization and Control 53
☆ Anderson Acceleration Without Restart: A Novel Method with $n$-Step Super Quadratic Convergence Rate
In this paper, we propose a novel Anderson's acceleration method to solve nonlinear equations, which does \emph{not} require a restart strategy to achieve numerical stability. We propose the greedy and random versions of our algorithm. Specifically, the greedy version selects the direction to maximize a certain measure of progress for approximating the current Jacobian matrix. In contrast, the random version chooses the random Gaussian vector as the direction to update the approximate Jacobian. Furthermore, our algorithm, including both greedy and random versions, has an $n$-step super quadratic convergence rate, where $n$ is the dimension of the objective problem. For example, the explicit convergence rate of the random version can be presented as $ \norm{\vx_{k+n+1} - \vx_*} / \norm{\vx_k- \vx_*}^2 = \cO\left(\left(1-\frac{1}{n}\right)^{kn}\right)$ for any $k\geq 0$ where $\vx_*$ is the optimum of the objective problem. This kind of convergence rate is new to Anderson's acceleration and quasi-Newton methods. The experiments also validate the fast convergence rate of our algorithm.
☆ SIS epidemics on open networks: A replacement-based approximation
In this paper we analyze continuous-time SIS epidemics subject to arrivals and departures of agents, by using an approximated process based on replacements. In defining the SIS dynamics in an open network, we consider a stochastic setting in which arrivals and departures take place according to Poisson processes with similar rates, and the new value of the infection probability of an arriving agent is drawn from a continuous distribution. Since the system size changes with time, we define an approximated process, in which replacements take place instead of arrivals and departures, and we focus on the evolution of an aggregate measure of the level of infection. So long as the reproduction number is less than one, the long-term behavior of this function measures the impact of the changes of the set of agents in the epidemic. We derive upper bounds for the expectation and variance of this function and we include a numerical example to show that the approximated process is close to the original SIS process.
comment: 7 pages, 2 figures, to appear in European Control Conference (ECC 2024)
Optimal Mass Transport of Nonlinear Systems under Input and Density Constraints
We investigate optimal mass transport problem of affine-nonlinear dynamical systems with input and density constraints. Three algorithms are proposed to tackle this problem, including two Uzawa-type methods and a splitting algorithm based on the Douglas-Rachford algorithm. Some preliminary simulation results are presented to demonstrate the effectiveness of our approaches.
☆ A Novel Loss Function-based Support Vector Machine for Binary Classification
The previous support vector machine(SVM) including $0/1$ loss SVM, hinge loss SVM, ramp loss SVM, truncated pinball loss SVM, and others, overlooked the degree of penalty for the correctly classified samples within the margin. This oversight affects the generalization ability of the SVM classifier to some extent. To address this limitation, from the perspective of confidence margin, we propose a novel Slide loss function ($\ell_s$) to construct the support vector machine classifier($\ell_s$-SVM). By introducing the concept of proximal stationary point, and utilizing the property of Lipschitz continuity, we derive the first-order optimality conditions for $\ell_s$-SVM. Based on this, we define the $\ell_s$ support vectors and working set of $\ell_s$-SVM. To efficiently handle $\ell_s$-SVM, we devise a fast alternating direction method of multipliers with the working set ($\ell_s$-ADMM), and provide the convergence analysis. The numerical experiments on real world datasets confirm the robustness and effectiveness of the proposed method.
☆ Improving the Optimization in Model Predictive Controllers: Scheduling Large Groups of Electric Vehicles
In parking lots with large groups of electric vehicles (EVs), charging has to happen in a coordinated manner, among others, due to the high load per vehicle and the limited capacity of the electricity grid. To achieve such coordination, model predictive control can be applied, thereby repeatedly solving an optimization problem. Due to its repetitive nature and its dependency on the time granularity, optimization has to be (computationally) efficient. The work presented here focuses on that optimization subroutine, its computational efficiency and how to speed up the optimization for large groups of EVs. In particular, we adapt FOCS, an algorithm that can solve the underlying optimization problem, to better suit the repetitive set-up of model predictive control by adding a pre-mature stop feature. Based on real-world data, we empirically show that the added feature speeds up the median computation time for 1-minute granularity by up to 44%. Furthermore, since FOCS is an algorithm that uses maximum flow methods as a subroutine, the impact of choosing various maximum flow methods on the runtime is investigated. Finally, we compare FOCS to a commercially available solver, concluding that FOCS outperforms the state-of-the-art when making a full-day schedule for large groups of EVs.
comment: 9 pages, 8 figures, latex template based on IEEE template https://www.overleaf.com/latex/templates/ieee-conference-template/grfzhhncsfqn
☆ A nonlocal approach to graded surface modeling in topology optimization
Additively manufactured structures often exhibit a correlation between their mechanical properties, such as stiffness, strength, and porosity, and their wall thickness. This correlation stems from the interplay between the manufacturing process and the properties of the filler material. In this study, we investigate the thickness-dependent effect on structural stiffness and propose a nonlocal integral model that introduces surface grading of Young's modulus to capture this phenomenon. We incorporate this model into topology optimization for designing structures with optimized compliance subject to a volume constraint. Notably, elastically degraded surfaces penalize excessively thin features, effectively eliminating them from the optimized design. We showcase the efficacy of our proposed framework by optimizing the design of a two-dimensional cantilever beam and a bridge.
☆ Identification of Cyclists' Route Choice Criteria
The behavior of cyclists when choosing the path to follow along a road network is not uniform. Some of them are mostly interested in minimizing the travelled distance, but some others may also take into account other features such as safety of the roads or pollution. Individuating the different groups of users, estimating the numerical consistency of each of these groups, and reporting the weights assigned by each group to different characteristics of the road network, is quite relevant. Indeed, when decision makers need to assign some budget for infrastructural interventions, they need to know the impact of their decisions, and this is strictly related to the way users perceive different features of the road network. In this paper, we propose an optimization approach to detect the weights assigned to different road features by various user groups, leveraging knowledge of the true paths followed by them, accessible, for example, through data collected by bike-sharing services.
☆ Extremality of collections of sets with respect to general perturbations
The paper proposes another extension of the extremal principle. A new extremality model involving arbitrary families of perturbations (deformations) of the given sets is studied. It generalizes the conventional model based on linear translations of the sets as well as its set-valued extensions. This approach leads to a more general and simpler version of fuzzy separation. We demonstrate the applicability of the new model to set-valued optimization problems, weakening the assumptions of the known results and streamlining their proofs.
comment: 29 pages
☆ Low-rank quaternion tensor completion for color video inpainting via a novel factorization strategy
Recently, a quaternion tensor product named Qt-product was proposed, and then the singular value decomposition and the rank of a third-order quaternion tensor were given. From a more applicable perspective, we extend the Qt-product and propose a novel multiplication principle for third-order quaternion tensor named gQt-product. With the gQt-product, we introduce a brand-new singular value decomposition for third-order quaternion tensors named gQt-SVD and then define gQt-rank and multi-gQt-rank. We prove that the optimal low-rank approximation of a third-order quaternion tensor exists and some numerical experiments demonstrate the low-rankness of color videos. So, we apply the low-rank quaternion tensor completion to color video inpainting problems and present alternating least-square algorithms to solve the proposed low gQt-rank and multi-gQt-rank quaternion tensor completion models. The convergence analyses of the proposed algorithms are established and some numerical experiments on various color video datasets show the high recovery accuracy and computational efficiency of our methods.
☆ Data-Driven Extrusion Force Control Tuning for 3D Printing
The quality of 3D prints often varies due to different conditions inherent to each print, such as filament type, print speed, and nozzle size. Closed-loop process control methods improve the accuracy and repeatability of 3D prints. However, optimal tuning of controllers for given process parameters and design geometry is often a challenge with manually tuned controllers resulting in inconsistent and suboptimal results. This work employs Bayesian optimization to identify the optimal controller parameters. Additionally, we explore transfer learning in the context of 3D printing by leveraging prior information from past trials. By integrating optimized extrusion force control and transfer learning, we provide a novel framework for closed-loop 3D printing and propose an automated calibration routine that produces high-quality prints for a desired combination of print settings, material, and shape.
comment: Submitted to IEEE CASE 2024
☆ Unbiased Extremum Seeking for PDEs IEEE Conference on Decision and Control
There have been recent efforts that combine seemingly disparate methods, extremum seeking (ES) optimization and partial differential equation (PDE) backstepping, to address the problem of model-free optimization with PDE actuator dynamics. In contrast to prior PDE-compensating ES designs, which only guarantee local stability around the extremum, we introduce unbiased ES that compensates for delay and diffusion PDE dynamics while ensuring exponential and unbiased convergence to the optimum. Our method leverages exponentially decaying/growing signals within the modulation/demodulation stages and carefully selected design parameters. The stability analysis of our designs relies on a state transformation, infinite-dimensional averaging, local exponential stability of the averaged system, local stability of the transformed system, and local exponential stability of the original system. Numerical simulations are presented to demonstrate the efficacy of the developed designs.
comment: Submitted to the 63rd IEEE Conference on Decision and Control (CDC), 2024
☆ Augmented Lagrangian method for coupled-cluster
We propose to improve the convergence properties of the single-reference coupled cluster (CC) method through an augmented Lagrangian formalism. The conventional CC method changes a linear high-dimensional eigenvalue problem with exponential size into a problem of determining the roots of a nonlinear system of equations that has a manageable size. However, current numerical procedures for solving this system of equations to get the lowest eigenvalue suffer from two practical issues: First, solving the CC equations may not converge, and second, when converging, they may converge to other -- potentially unphysical -- states, which are stationary points of the CC energy expression. We show that both issues can be dealt with when a suitably defined energy is minimized in addition to solving the original CC equations. We further propose an augmented Lagrangian method for coupled cluster (alm-CC) to solve the resulting constrained optimization problem. We numerically investigate the proposed augmented Lagrangian formulation showing that the convergence towards the ground state is significantly more stable and that the optimization procedure is less susceptible to local minima. Furthermore, the computational cost of alm-CC is comparable to the conventional CC method.
comment: 17 pages, 8 figures
☆ On solution of tropical discrete best approximation problems
We consider a discrete best approximation problem formulated in the framework of tropical algebra, which deals with the theory and applications of algebraic systems with idempotent operations. Given a set of samples of input and output of an unknown function, the problem is to construct a generalized tropical Puiseux polynomial that best approximates the function in the sense of a tropical distance function. The construction of an approximate polynomial involves the evaluation of both unknown coefficient and exponent of each monomial in the polynomial. To solve the approximation problem, we first reduce the problem to an equation in unknown vector of coefficients, which is given by a matrix with entries parameterized by unknown exponents. We derive a best approximate solution of the equation, which yields both vector of coefficients and approximation error parameterized by the exponents. Optimal values of exponents are found by minimization of the approximation error, which is reduced to a minimization of a function of exponents over all partitions of a finite set. We solve this minimization problem in terms of max-plus algebra (where addition is defined as maximum and multiplication as arithmetic addition) by using a computational procedure based on the agglomerative clustering technique. This solution is extended to the minimization problem of finding optimal exponents in the polynomial in terms of max-algebra (where addition is defined as maximum). The results obtained are applied to develop new solutions for conventional problems of discrete best approximation of real functions by piecewise linear functions and piecewise Puiseux polynomials. We discuss computational complexity of the proposed solution and estimate upper bounds on the computational time. We demonstrate examples of approximation problems solved in terms of max-plus and max-algebra, and give graphical illustrations.
comment: 21 pages, 6 figures
☆ DASA: Delay-Adaptive Multi-Agent Stochastic Approximation
We consider a setting in which $N$ agents aim to speedup a common Stochastic Approximation (SA) problem by acting in parallel and communicating with a central server. We assume that the up-link transmissions to the server are subject to asynchronous and potentially unbounded time-varying delays. To mitigate the effect of delays and stragglers while reaping the benefits of distributed computation, we propose \texttt{DASA}, a Delay-Adaptive algorithm for multi-agent Stochastic Approximation. We provide a finite-time analysis of \texttt{DASA} assuming that the agents' stochastic observation processes are independent Markov chains. Significantly advancing existing results, \texttt{DASA} is the first algorithm whose convergence rate depends only on the mixing time $\tmix$ and on the average delay $\tau_{avg}$ while jointly achieving an $N$-fold convergence speedup under Markovian sampling. Our work is relevant for various SA applications, including multi-agent and distributed temporal difference (TD) learning, Q-learning and stochastic optimization with correlated data.
☆ Tightness of the matrix Moment-SOS hierarchy
This paper studies the matrix Moment-SOS hierarchy for solving polynomial matrix optimization. Our first result is to show the tightness (i.e., the finite convergence) of this hierarchy, if the nondegeneracy condition, strict complementarity condition and second order sufficient condition hold at every minimizer, under the usual archimedeanness assumption. A useful criterion for detecting tightness is the flat truncation. Our second result is to show that every minimizer of the moment relaxation must have a flat truncation when the relaxation order is big enough, under the above mentioned optimality assumptions. These results give connections between nonlinear semidefinite optimization theory and Moment-SOS methods for solving polynomial matrix optimization.
☆ Robust Finite-time Stabilization of Linear Systems with Limited State Quantization
This paper investigates the robust asymptotic stabilization of a linear time-invariant (LTI) system by a static feedback with a static state quantization. It is shown that the controllable LTI system can be stabilized to zero in a finite time by means of a nonlinear feedback with a quantizer having a limited (finite) number of values (quantization seeds) even when all parameters of the controller and the quantizer are time-invariant. The control design is based on generalized homogeneity. A homogeneous spherical quantizer is introduced. The static homogeneous feedback is shown to be local (or global) finite-time stabilizer for the linear system (dependently of the system matrix). The tuning rules for both the quantizer and the feedback law are obtained in the form of Linear Matrix Inequalities (LMIs). The closed-loop system is proven to be robust with respect to some bounded matched and vanishing mismatched perturbations. Theoretical results are supported by numerical simulations. \
☆ Belief Samples Are All You Need For Social Learning
In this paper, we consider the problem of social learning, where a group of agents embedded in a social network are interested in learning an underlying state of the world. Agents have incomplete, noisy, and heterogeneous sources of information, providing them with recurring private observations of the underlying state of the world. Agents can share their learning experience with their peers by taking actions observable to them, with values from a finite feasible set of states. Actions can be interpreted as samples from the beliefs which agents may form and update on what the true state of the world is. Sharing samples, in place of full beliefs, is motivated by the limited communication, cognitive, and information-processing resources available to agents especially in large populations. Previous work (Salhab et al.) poses the question as to whether learning with probability one is still achievable if agents are only allowed to communicate samples from their beliefs. We provide a definite positive answer to this question, assuming a strongly connected network and a ``collective distinguishability'' assumption, which are both required for learning even in full-belief-sharing settings. In our proposed belief update mechanism, each agent's belief is a normalized weighted geometric interpolation between a fully Bayesian private belief -- aggregating information from the private source -- and an ensemble of empirical distributions of the samples shared by her neighbors over time. By carefully constructing asymptotic almost-sure lower/upper bounds on the frequency of shared samples matching the true state/or not, we rigorously prove the convergence of all the beliefs to the true state, with probability one.
comment: 6 pages
☆ Output-feedback Synthesis Orbit Geometry: Quotient Manifolds and LQG Direct Policy Optimization
In this paper, we consider direct policy optimization for the linear-quadratic Gaussian (LQG) setting. Over the past few years, it has been recognized that the landscape of stabilizing output-feedback controllers of relevance to LQG has an intricate geometry, particularly as it pertains to the existence of spurious stationary points. In order to address such challenges, in this paper, we first adopt a Riemannian metric for the space of stabilizing full-order minimal output-feedback controllers. We then proceed to prove that the orbit of such controllers modulo coordinate transformation admits a Riemannian quotient manifold structure. This geometric structure is then used to develop a Riemannian gradient descent for the direct LQG policy optimization. We prove a local convergence guarantee with linear rate and show the proposed approach exhibits significantly faster and more robust numerical performance as compared with ordinary gradient descent for LQG. Subsequently, we provide reasons for this observed behavior; in particular, we argue that optimizing over the orbit space of controllers is the right theoretical and computational setup for direct LQG policy optimization.
☆ Approximation with Random Shallow ReLU Networks with Applications to Model Reference Adaptive Control
Neural networks are regularly employed in adaptive control of nonlinear systems and related methods o reinforcement learning. A common architecture uses a neural network with a single hidden layer (i.e. a shallow network), in which the weights and biases are fixed in advance and only the output layer is trained. While classical results show that there exist neural networks of this type that can approximate arbitrary continuous functions over bounded regions, they are non-constructive, and the networks used in practice have no approximation guarantees. Thus, the approximation properties required for control with neural networks are assumed, rather than proved. In this paper, we aim to fill this gap by showing that for sufficiently smooth functions, ReLU networks with randomly generated weights and biases achieve $L_{\infty}$ error of $O(m^{-1/2})$ with high probability, where $m$ is the number of neurons. It suffices to generate the weights uniformly over a sphere and the biases uniformly over an interval. We show how the result can be used to get approximations of required accuracy in a model reference adaptive control application.
comment: Under Review for Conference on Decision and Control
☆ Practical Acceleration of the Condat-Vũ Algorithm
The Condat-V\~u algorithm is a widely used primal-dual method for optimizing composite objectives of three functions. Several algorithms for optimizing composite objectives of two functions are special cases of Condat-V\~u, including proximal gradient descent (PGD). It is well-known that PGD exhibits suboptimal performance, and a simple adjustment to PGD can accelerate its convergence rate from $\mathcal{O}(1/T)$ to $\mathcal{O}(1/T^2)$ on convex objectives, and this accelerated rate is optimal. In this work, we show that a simple adjustment to the Condat-V\~u algorithm allows it to recover accelerated PGD (APGD) as a special case, instead of PGD. We prove that this accelerated Condat--V\~u algorithm achieves optimal convergence rates and significantly outperforms the traditional Condat-V\~u algorithm in regimes where the Condat--V\~u algorithm approximates the dynamics of PGD. We demonstrate the effectiveness of our approach in various applications in machine learning and computational imaging.
☆ Non-stationary Bandits with Habituation and Recover Dynamics and Knapsack Constraints
Multi-armed bandit models have proven to be useful in modeling many real world problems in the areas of control and sequential decision making with partial information. However, in many scenarios, such as those prevalent in healthcare and operations management, the decision maker's expected reward will decrease if an action is selected too frequently while it may recover if they abstain from selecting this action. This scenario is further complicated when choosing a particular action also expends a random amount of a limited resource where the distribution is also initially unknown to the decision maker. In this paper we study a class of models that address this setting that we call reducing or gaining unknown efficacy bandits with stochastic knapsack constraints (ROGUEwK). We propose a combination upper confidence bound (UCB) and lower confidence bound (LCB) approximation algorithm for optimizing this model. Our algorithm chooses which action to play at each time point by solving a linear program (LP) with the UCB for the average rewards and LCB for the average costs as inputs. We show that the regret of our algorithm is sub-linear as a function of time and total constraint budget when compared to a dynamic oracle. We validate the performance of our algorithm against existing state of the art non-stationary and knapsack bandit approaches in a simulation study and show that our methods are able to on average achieve a 13% improvement in terms of total reward.
comment: 8 pages, 3 figures
☆ State-Augmented Linear Games with Antagonistic Error for High-Dimensional, Nonlinear Hamilton-Jacobi Reachability
Hamilton-Jacobi Reachability (HJR) is a popular method for analyzing the liveness and safety of a dynamical system with bounded control and disturbance. The corresponding HJ value function offers a robust controller and characterizes the reachable sets, but is traditionally solved with Dynamic Programming (DP) and limited to systems of dimension less than six. Recently, the space-parallelizeable, generalized Hopf formula has been shown to also solve the HJ value with a nearly three-log increase in dimension limit, but is limited to linear systems. To extend this potential, we demonstrate how state-augmented (SA) spaces, which are well-known for their improved linearization accuracy, may be used to solve tighter, conservative approximations of the value function with any linear model in this SA space. Namely, we show that with a representation of the true dynamics in the SA space, a series of inequalities confirms that the value of a SA linear game with antagonistic error is a conservative envelope of the true value function. It follows that if the optimal controller for the HJ SA linear game with error may succeed, it will also succeed in the true system. Unlike previous methods, this result offers the ability to safely approximate reachable sets and their corresponding controllers with the Hopf formula in a non-convex manner. Finally, we demonstrate this in the slow manifold system for clarity, and in the controlled Van der Pol system with different lifting functions.
☆ An Optimal Solution to Infinite Horizon Nonlinear Control Problems: Part II
This paper considers the infinite horizon optimal control problem for nonlinear systems. Under the condition of nonlinear controllability of the system to any terminal set containing the origin and forward invariance of the terminal set, we establish a regularized solution approach consisting of a ``finite free final time" optimal transfer problem to the terminal set which renders the set globally asymptotically stable. Further, we show that the approximations converge to the optimal infinite horizon cost as the size of the terminal set decreases to zero. We also perform the analysis for the discounted problem and show that the terminal set is asymptotically stable only for a subset of the state space and not globally. The theory is empirically evaluated on various nonholonomic robotic systems to show that the cost of our approximate problem converges and the transfer time into the terminal set is dependent on the initial state of the system, necessitating the free final time formulation.
☆ Unsupervised Feature Selection via Nonnegative Orthogonal Constrained Regularized Minimization
Unsupervised feature selection has drawn wide attention in the era of big data since it is a primary technique for dimensionality reduction. However, many existing unsupervised feature selection models and solution methods were presented for the purpose of application, and lack of theoretical support, e.g., without convergence analysis. In this paper, we first establish a novel unsupervised feature selection model based on regularized minimization with nonnegative orthogonal constraints, which has advantages of embedding feature selection into the nonnegative spectral clustering and preventing overfitting. An effective inexact augmented Lagrangian multiplier method is proposed to solve our model, which adopts the proximal alternating minimization method to solve subproblem at each iteration. We show that the sequence generated by our method globally converges to a Karush-Kuhn-Tucker point of our model. Extensive numerical experiments on popular datasets demonstrate the stability and robustness of our method. Moreover, comparison results of algorithm performance show that our method outperforms some existing state-of-the-art methods.
☆ An $α$-potential game framework for $N$-player games
This paper proposes and studies a general form of dynamic $N$-player non-cooperative games called $\alpha$-potential games, where the change of a player's value function upon her unilateral deviation from her strategy is equal to the change of an $\alpha$-potential function up to an error $\alpha$. Analogous to the static potential game (which corresponds to $\alpha=0$), the $\alpha$-potential game framework is shown to reduce the challenging task of finding approximate Nash equilibria for a dynamic game to minimizing the $\alpha$-potential function. Moreover, an analytical characterization of $\alpha$-potential functions is established, with $\alpha$ represented in terms of the magnitude of the asymmetry of value functions' second-order derivatives. For stochastic differential games in which the state dynamic is a controlled diffusion, $\alpha$ is explicitly identified in terms of the number of players, the choice of admissible strategies, and the intensity of interactions, and the level of heterogeneity among players. This is achieved by introducing a suitable linear derivative of the value functions with respect to unilateral deviations of strategies and via analyzing the sensitivity processes of state dynamics with respect to controls. For games with mean-field type interactions, $\alpha$ is shown to decay to zero as the number of players goes to infinity, even with heterogeneity in state dynamics, cost functions, and admissible strategy classes. For distributed games, if a static potential function can be derived from the cost functions, then $\alpha=0$. For crowd aversion games, $\alpha$ is capable of capturing the subtle difference between the choice of admissible strategies.
☆ Provably Robust Score-Based Diffusion Posterior Sampling for Plug-and-Play Image Reconstruction
In a great number of tasks in science and engineering, the goal is to infer an unknown image from a small number of measurements collected from a known forward model describing certain sensing or imaging modality. Due to resource constraints, this task is often extremely ill-posed, which necessitates the adoption of expressive prior information to regularize the solution space. Score-based diffusion models, due to its impressive empirical success, have emerged as an appealing candidate of an expressive prior in image reconstruction. In order to accommodate diverse tasks at once, it is of great interest to develop efficient, consistent and robust algorithms that incorporate {\em unconditional} score functions of an image prior distribution in conjunction with flexible choices of forward models. This work develops an algorithmic framework for employing score-based diffusion models as an expressive data prior in general nonlinear inverse problems. Motivated by the plug-and-play framework in the imaging community, we introduce a diffusion plug-and-play method (\textsf{DPnP}) that alternatively calls two samplers, a proximal consistency sampler based solely on the likelihood function of the forward model, and a denoising diffusion sampler based solely on the score functions of the image prior. The key insight is that denoising under white Gaussian noise can be solved {\em rigorously} via both stochastic (i.e., DDPM-type) and deterministic (i.e., DDIM-type) samplers using the unconditional score functions. We establish both asymptotic and non-asymptotic performance guarantees of \textsf{DPnP}, and provide numerical experiments to illustrate its promise in solving both linear and nonlinear image reconstruction tasks. To the best of our knowledge, \textsf{DPnP} is the first provably-robust posterior sampling method for nonlinear inverse problems using unconditional diffusion priors.
☆ Algebraic Constraints on Common Lines with Applications to Community Detection in Cryo-EM
We revisit the topic of common lines between 2D class averages in single particle cryo-electron microscopy (cryo-EM). We derive a novel low-rank constraint on a certain $2n \times n$ matrix storing properly-scaled basis vectors for the common lines between $n$ tomographic images of one molecular conformation. Using this constraint and others, we introduce an optimization algorithm to denoise the common lines of one conformation. As a main application, we develop a clustering algorithm to partition a set of noisy common lines into homogeneous communities, in the case of discrete heterogeneity in cryo-EM. We demonstrate the methods on synthetic and experimental datasets.
Optimality of spherical codes via exact semidefinite programming bounds
We show that the spectral embeddings of all known triangle-free strongly regular graphs are optimal spherical codes (the new cases are $56$ points in $20$ dimensions, $50$ points in $21$ dimensions, and $77$ points in $21$ dimensions), as are certain mutually unbiased basis arrangements constructed using Kerdock codes in up to $1024$ dimensions (namely, $2^{4k} + 2^{2k+1}$ points in $2^{2k}$ dimensions for $2 \le k \le 5$). As a consequence of the latter, we obtain optimality of the Kerdock binary codes of block length $64$, $256$, and $1024$, as well as uniqueness for block length $64$. We also prove universal optimality for $288$ points on a sphere in $16$ dimensions. To prove these results, we use three-point semidefinite programming bounds, for which only a few sharp cases were known previously. To obtain rigorous results, we develop improved techniques for rounding approximate solutions of semidefinite programs to produce exact optimal solutions.
comment: 32 pages, 1 figure
☆ Improved convergence rates for the Difference-of-Convex algorithm
We consider a difference-of-convex formulation where one of the terms is allowed to be hypoconvex (or weakly convex). We first examine the precise behavior of a single iteration of the Difference-of-Convex algorithm (DCA), giving a tight characterization of the objective function decrease. This requires distinguishing between eight distinct parameter regimes. Our proofs are inspired by the performance estimation framework, but are much simplified compared to similar previous work. We then derive sublinear DCA convergence rates towards critical points, distinguishing between cases where at least one of the functions is smooth and where both functions are nonsmooth. We conjecture the tightness of these rates for four parameter regimes, based on strong numerical evidence obtained via performance estimation, as well as the leading constant in the asymptotic sublinear rate for two more regimes.
☆ Weak Convergence Analysis of Online Neural Actor-Critic Algorithms
We prove that a single-layer neural network trained with the online actor critic algorithm converges in distribution to a random ordinary differential equation (ODE) as the number of hidden units and the number of training steps $\rightarrow \infty$. In the online actor-critic algorithm, the distribution of the data samples dynamically changes as the model is updated, which is a key challenge for any convergence analysis. We establish the geometric ergodicity of the data samples under a fixed actor policy. Then, using a Poisson equation, we prove that the fluctuations of the model updates around the limit distribution due to the randomly-arriving data samples vanish as the number of parameter updates $\rightarrow \infty$. Using the Poisson equation and weak convergence techniques, we prove that the actor neural network and critic neural network converge to the solutions of a system of ODEs with random initial conditions. Analysis of the limit ODE shows that the limit critic network will converge to the true value function, which will provide the actor an asymptotically unbiased estimate of the policy gradient. We then prove that the limit actor network will converge to a stationary point.
☆ Local search and trajectory metaheuristics for the flexible job shop scheduling problem with sequencing flexibility and position-based learning effect
The flexible job shop scheduling problem with sequencing flexibility and position-based learning effect is considered in the present work. In [K. A. G. Araujo, E. G. Birgin, and D. P. Ronconi, Technical Report MCDO02022024, 2024], models, constructive heuristics, and benchmark instances for the same problem were introduced. In the present work, we are concerned with the development of effective and efficient methods for its resolution. For this purpose, a local search method and four trajectory metaheuristics are considered. In the local search, we show that the classical strategy of only reallocating operations that are part of the critical path can miss better quality neighbors, as opposed to what happens in the case where there is no learning effect. Consequently, we analyze an alternative type of neighborhood reduction that eliminates only neighbors that are not better than the current solution. In addition, we also suggest a neighborhood cut and experimentally verify that this significantly reduces the neighborhood size, bringing efficiency, with minimal loss in effectiveness. Extensive numerical experiments with the local search and the metaheuristics are carried on. The experiments show that tabu search, built on the reduced neighborhood, when applied to large-sized instances, stands out in relation to other the other three metaheuristics, namely, iterated local search, greedy randomized adaptive search procedure, and simulating annealing. Experiments with classical instances without sequencing flexibility show that the introduced methods also stand out in relation to methods from the literature. All the methods introduced, as well as the instances and solutions found, are freely available. As a whole, we build a test suite that can be used in future work.
☆ Stochastic Inertial Dynamics Via Time Scaling and Averaging
Our work is part of the close link between continuous-time dissipative dynamical systems and optimization algorithms, and more precisely here, in the stochastic setting. We aim to study stochastic convex minimization problems through the lens of stochastic inertial differential inclusions that are driven by the subgradient of a convex objective function. This will provide a general mathematical framework for analyzing the convergence properties of stochastic second-order inertial continuous-time dynamics involving vanishing viscous damping and measurable stochastic subgradient selections. Our chief goal in this paper is to develop a systematic and unified way that transfers the properties recently studied for first-order stochastic differential equations to second-order ones involving even subgradients in lieu of gradients. This program will rely on two tenets: time scaling and averaging, following an approach recently developed in the literature by one of the co-authors in the deterministic case. Under a mild integrability assumption involving the diffusion term and the viscous damping, our first main result shows that almost surely, there is weak convergence of the trajectory towards a minimizer of the objective function and fast convergence of the values and gradients. We also provide a comprehensive complexity analysis by establishing several new pointwise and ergodic convergence rates in expectation for the convex, strongly convex, and (local) Polyak-Lojasiewicz case. Finally, using Tikhonov regularization with a properly tuned vanishing parameter, we can obtain almost sure strong convergence of the trajectory towards the minimum norm solution.
comment: 31 pages. arXiv admin note: text overlap with arXiv:2403.06708
☆ Models, constructive heuristics, and benchmark instances for the flexible job shop scheduling problem with sequencing flexibility and position-based learning effect
This paper addresses the flexible job shop scheduling problem with sequencing flexibility and position-based learning effect. In this variant of the flexible job shop scheduling problem, precedence constraints of the operations constituting a job are given by an arbitrary directed acyclic graph, in opposition to the classical case in which a total order is imposed. Additionally, it is assumed that the processing time of an operation in a machine is subject to a learning process such that the larger the position of the operation in the machine, the faster the operation is processed. Mixed integer programming and constraint programming models are presented and compared in the present work. In addition, constructive heuristics are introduced to provide an initial solution to the models' solvers. Sets of benchmark instances are also introduced. The problem considered corresponds to modern problems of great relevance in the printing industry. The models and instances presented are intended to support the development of new heuristic and metaheuristics methods for this problem.
☆ The stability of the multivariate geometric Brownian motion as a bilinear matrix inequality problem
In this manuscript, we study the stability of the origin for the multivariate geometric Brownian motion. More precisely, under suitable sufficient conditions, we construct a Lyapunov function such that the origin of the multivariate geometric Brownian motion is globally asymptotically stable in probability. Moreover, we show that such conditions can be rewritten as a Bilinear Matrix Inequality (BMI) feasibility problem. We stress that no commutativity relations between the drift matrix and the noise dispersion matrices are assumed and therefore the so-called Magnus representation of the solution of the multivariate geometric Brownian motion is complicated. In addition, we exemplify our method in numerous specific models from the literature such as random linear oscillators, satellite dynamics, inertia systems, diagonal noise systems, cancer self-remission and smoking.
comment: 34 pages
♻ ☆ Combining Learning and Control in Linear Systems
In this paper, we provide a theoretical framework that separates the control and learning tasks in a linear system. This separation allows us to combine offline model-based control with online learning approaches and thus circumvent current challenges in deriving optimal control strategies in applications where a large volume of data is added to the system gradually in real time and not altogether in advance. We provide an analytical example to illustrate the framework.
comment: 6 pages, 1 figure. arXiv admin note: text overlap with arXiv:2211.14972
♻ ☆ Peak Estimation of Rational Systems using Convex Optimization
This paper presents algorithms that upper-bound the peak value of a state function along trajectories of a continuous-time system with rational dynamics. The finite-dimensional but nonconvex peak estimation problem is cast as a convex infinite-dimensional linear program in occupation measures. This infinite-dimensional program is then truncated into finite-dimensions using the moment-Sum-of-Squares (SOS) hierarchy of semidefinite programs. Prior work on treating rational dynamics using the moment-SOS approach involves clearing dynamics to common denominators or adding lifting variables to handle reciprocal terms under new equality constraints. Our solution method uses a sum-of-rational method based on absolute continuity of measures. The Moment-SOS truncations of our program possess lower computational complexity and (empirically demonstrated) higher accuracy of upper bounds on example systems as compared to prior approaches.
comment: 9 pages, 2 figures, 4 tables
♻ ☆ An equivalent reformulation and multi-proximity gradient algorithms for a class of nonsmooth fractional programming
In this paper, we consider a class of structured fractional programs, where the numerator part is the sum of a block-separable (possibly nonsmooth nonconvex) function and a locally Lipschitz differentiable (possibly nonconvex) function, while the denominator is a convex (possibly nonsmooth) function. We first present a novel reformulation for the original problem and show the relationship between optimal solutions, critical points and KL exponents of these two problems. Inspired by the reformulation, we propose a flexible framework of multi-proximity gradient algorithms (MPGA), which computes the proximity operator with respect to the Fenchel conjugate associated with the convex denominator of the original problem rather than evaluating its subgradient as in the existing methods. Also, MPGA employs a nonmonotone linear-search scheme in its gradient descent step, since the smooth part in the numerator of the original problem is not globally Lipschitz differentiable. Based on the framework of MPGA, we develop two specific algorithms, namely, cyclic MPGA and randomized MPGA, and establish their subsequential convergence under mild conditions. Moreover, the sequential convergence of cyclic MPGA with the monotone line-search (CMPGA_ML) is guaranteed if the extended objective associated with the reformulated problem satisfies the Kurdyka-{\L}ojasiewicz (KL) property and some other mild assumptions. In particular, we prove that the corresponding KL exponents are 1/2 for several special cases of the fractional programs, and so, CMPGA_ML exhibits a linear convergence rate. Finally, some preliminary numerical experiments are performed to demonstrate the efficiency of our proposed algorithms.
♻ ☆ Markovian lifting and optimal control for integral stochastic Volterra equations with completely monotone kernels
In this paper, we focus on solving the optimal control problem for integral stochastic Volterra equations in a finite dimensional setting. In our setting, the noise term is driven by a pure jump L\'evy noise and the control acts on the intensity of the jumps. We use recent techniques proposed by Hamaguchi, where a crucial requirement is that the convolution kernel should be a completely monotone function. This allows us to use Bernstein's representation and the machinery of Laplace transform to obtain a Markovian lift. It is natural that the Markovian lift, in whatever form constructed, transforms the state equation into a stochastic differential equation in an infinite-dimensional space. This space should be large enough to contain all the information about the history of the process. Hence, although the original equation is taken in a finite dimensional space, the resulting lift is always infinite dimensional. We solve the problem by using the forward-backward approach in the infinite-dimensional setting and prove the existence of the optimal control for the original problem. Under additional assumptions on the coefficients, we see that a control in closed-loop form can be achieved.
comment: 19 pages. v2: updated references
♻ ☆ A Low-Rank ADMM Splitting Approach for Semidefinite Programming
We introduce a new first-order method for solving general semidefinite programming problems, based on the alternating direction method of multipliers (ADMM) and a matrix-splitting technique. Our algorithm has an advantage over the Burer-Monteiro approach as it only involves much easier quadratically regularized subproblems in each iteration. For a linear objective, the subproblems are well-conditioned quadratic programs that can be efficiently solved by the standard conjugate gradient method. We show that the ADMM algorithm achieves sublinear or linear convergence rates to the KKT solutions under different conditions. Building on this theoretical development, we present LoRADS, a new solver for linear SDP based on the Low-Rank ADMM Splitting approach. LoRADS incorporates several strategies that significantly increase its efficiency. Firstly, it initiates with a warm-start phase that uses the Burer-Monteiro approach. Moreover, motivated by the SDP low-rank theory [So et al. 2008], LoRADS chooses an initial rank of logarithmic order and then employs a dynamic approach to increase the rank. Numerical experiments indicate that LoRADS exhibits promising performance on various SDP problems. A noteworthy achievement of LoRADS is its successful solving of a matrix completion problem with $15,694,167$ constraints and a matrix variable of size $40,000 \times 40,000$ in $351$ seconds.
♻ ☆ PowerSimulationsDynamics.jl -- An Open Source Modeling Package for Modern Power Systems with Inverter-Based Resources
In this paper we present the development of an open-source simulation toolbox, PowerSimulationsDynamics.jl, to study the dynamic response of power systems, focusing on the requirements to model systems with high penetrations of Inverter-Based Resources (IBRs). PowerSimulationsDynamics.jl is implemented in Julia and features a rich library of synchronous generator, inverter, and load models. In addition, it allows the study of quasi-static phasors and electromagnetic dq models that use a dynamic network representation. Case studies and validation exercises show that PowerSimulationsDynamics.jl results closely match other commercial and open-source simulation tools.
♻ ☆ Stochastic Approximation with Delayed Updates: Finite-Time Rates under Markovian Sampling AISTATS
Motivated by applications in large-scale and multi-agent reinforcement learning, we study the non-asymptotic performance of stochastic approximation (SA) schemes with delayed updates under Markovian sampling. While the effect of delays has been extensively studied for optimization, the manner in which they interact with the underlying Markov process to shape the finite-time performance of SA remains poorly understood. In this context, our first main contribution is to show that under time-varying bounded delays, the delayed SA update rule guarantees exponentially fast convergence of the \emph{last iterate} to a ball around the SA operator's fixed point. Notably, our bound is \emph{tight} in its dependence on both the maximum delay $\tau_{max}$, and the mixing time $\tau_{mix}$. To achieve this tight bound, we develop a novel inductive proof technique that, unlike various existing delayed-optimization analyses, relies on establishing uniform boundedness of the iterates. As such, our proof may be of independent interest. Next, to mitigate the impact of the maximum delay on the convergence rate, we provide the first finite-time analysis of a delay-adaptive SA scheme under Markovian sampling. In particular, we show that the exponent of convergence of this scheme gets scaled down by $\tau_{avg}$, as opposed to $\tau_{max}$ for the vanilla delayed SA rule; here, $\tau_{avg}$ denotes the average delay across all iterations. Moreover, the adaptive scheme requires no prior knowledge of the delay sequence for step-size tuning. Our theoretical findings shed light on the finite-time effects of delays for a broad class of algorithms, including TD learning, Q-learning, and stochastic gradient descent under Markovian sampling.
comment: Accepted to the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024!
♻ ☆ Federated Learning Using Three-Operator ADMM
Federated learning (FL) has emerged as an instance of distributed machine learning paradigm that avoids the transmission of data generated on the users' side. Although data are not transmitted, edge devices have to deal with limited communication bandwidths, data heterogeneity, and straggler effects due to the limited computational resources of users' devices. A prominent approach to overcome such difficulties is FedADMM, which is based on the classical two-operator consensus alternating direction method of multipliers (ADMM). The common assumption of FL algorithms, including FedADMM, is that they learn a global model using data only on the users' side and not on the edge server. However, in edge learning, the server is expected to be near the base station and have direct access to rich datasets. In this paper, we argue that leveraging the rich data on the edge server is much more beneficial than utilizing only user datasets. Specifically, we show that the mere application of FL with an additional virtual user node representing the data on the edge server is inefficient. We propose FedTOP-ADMM, which generalizes FedADMM and is based on a three-operator ADMM-type technique that exploits a smooth cost function on the edge server to learn a global model parallel to the edge devices. Our numerical experiments indicate that FedTOP-ADMM has substantial gain up to 33\% in communication efficiency to reach a desired test accuracy with respect to FedADMM, including a virtual user on the edge server.
comment: accepted to IEEE Journal of Selected Topics in Signal Processing, 2022
♻ ☆ Multilinear formulations for computing Nash equilibrium of multi-player matrix games
We present multilinear and mixed-integer multilinear programs to find a Nash equilibrium in multi-player noncooperative games. We compare the formulations to common algorithms in Gambit, and conclude that a multilinear feasibility program finds a Nash equilibrium faster than any of the methods we compare it to, including the quantal response equilibrium method, which is recommended for large games. Hence, the multilinear feasibility program is an alternative method to find a Nash equilibrium in multi-player games, and outperforms many common algorithms. The mixed-integer formulations are generalisations of known mixed-integer programs for two-player games, however unlike two-player games, these mixed-integer programs do not give better performance than existing algorithms.
comment: 15 page conference paper accepted
♻ ☆ On the Diameter of a 2-Sum of Polyhedra
The study of the combinatorial diameter of a polyhedron is a classical topic in linear-programming theory due to its close connection with the possibility of a polynomial simplex-method pivot rule. The 2-sum operation is a classical operation for graphs, matrices, and matroids; we extend this definition to polyhedra. We analyze the diameters of 2-sum polyhedra, which are those polyhedra that arise from this operation. These polyhedra appear in matroid and integer-programming theory as a natural way to link two systems in a joint model with a single shared constraint and the 2-sum also appears as a key operation in Seymour's decomposition theorem for totally-unimodular matrices. We show that the diameter of a 2-sum polyhedron is quadratic in the diameters of its summands. The methods transfer to a linear bound for the addition of a unit column to an equality system, or equivalently, to the relaxation of an equality constraint to an inequality constraint. Further, we use our methods to analyze the distance between vertices on certain faces of a 3-sum polyhedron.
♻ ☆ Long Solution Times or Low Solution Quality: On Trade-Offs in Choosing a Power Flow Formulation for the Optimal Power Shutoff Problem
The Optimal Power Shutoff (OPS) problem is an optimization problem that makes power line de-energization decisions in order to reduce the risk of igniting a wildfire, while minimizing the load shed of customers. This problem, with DC linear power flow equations, has been used in many studies in recent years. However, using linear approximations for power flow when making decisions on the network topology is known to cause challenges with AC feasibility of the resulting network, as studied in the related contexts of optimal transmission switching or grid restoration planning. This paper explores the accuracy of the DC OPS formulation and the ability to recover an AC-feasible power flow solution after de-energization decisions are made. We also extend the OPS problem to include variants with the AC, Second-Order-Cone, and Network-Flow power flow equations, and compare them to the DC approximation with respect to solution quality and time. The results highlight that the DC approximation overestimates the amount of load that can be served, leading to poor de-energization decisions. The AC and SOC-based formulations are better, but prohibitively slow to solve for even modestly sized networks thus demonstrating the need for new solution methods with better trade-offs between computational time and solution quality.
♻ ☆ A distribution-free mixed-integer optimization approach to hierarchical modelling of clustered and longitudinal data
Recent advancements in Mixed Integer Optimization (MIO) algorithms, paired with hardware enhancements, have led to significant speedups in resolving MIO problems. These strategies have been utilized for optimal subset selection, specifically for choosing $k$ features out of $p$ in linear regression given $n$ observations. In this paper, we broaden this method to facilitate cluster-aware regression, where selection aims to choose $\lambda$ out of $K$ clusters in a linear mixed effects (LMM) model with $n_k$ observations for each cluster. Through comprehensive testing on a multitude of synthetic and real datasets, we exhibit that our method efficiently solves problems within minutes. Through numerical experiments, we also show that the MIO approach outperforms both Gaussian- and Laplace-distributed LMMs in terms of generating sparse solutions with high predictive power. Traditional LMMs typically assume that clustering effects are independent of individual features. However, we introduce an innovative algorithm that evaluates cluster effects for new data points, thereby increasing the robustness and precision of this model. The inferential and predictive efficacy of this approach is further illustrated through its application in student scoring and protein expression.
♻ ☆ A Tikhonov theorem for McKean-Vlasov two-scale systems and a new application to mean field optimal control problems
We provide a new version of the Tikhonov theorem for both two-scale forward systems and also two-scale forward-backward systems of stochastic differential equations, which also covers the McKean-Vlasov case. Differently from what is usually done in the literature, we prove a type of convergence for the ''fast'' variable, which allows the limiting process to be discontinuous. This is relevant for the second part of the paper, where we present a new application of this theory to the approximation of the solution of mean field control problems. Towards this aim, we construct a two-scale system whose ''fast'' component converges to the optimal control process, while the ''slow'' component converges to the optimal state process. The interest in such a procedure is that it allows to approximate the solution of the control problem avoiding the usual step of the minimization of the Hamiltonian.
♻ ☆ Decentralized Finite-Sum Optimization over Time-Varying Networks
We consider decentralized time-varying stochastic optimization problems where each of the functions held by the nodes has a finite sum structure. Such problems can be efficiently solved using variance reduction techniques. Our aim is to explore the lower complexity bounds (for communication and number of stochastic oracle calls) and find optimal algorithms. The paper studies strongly convex and nonconvex scenarios. To the best of our knowledge, variance reduced schemes and lower bounds for time-varying graphs have not been studied in the literature. For nonconvex objectives, we obtain lower bounds and develop an optimal method GT-PAGE. For strongly convex objectives, we propose the first decentralized time-varying variance-reduction method ADOM+VR and establish lower bound in this scenario, highlighting the open question of matching the algorithms complexity and lower bounds even in static network case.
♻ ☆ Hybrid physics-informed metabolic cybergenetics: process rates augmented with machine-learning surrogates informed by flux balance analysis
Metabolic cybergenetics is a promising concept that interfaces gene expression and cellular metabolism with computers for real-time dynamic metabolic control. The focus is on control at the transcriptional level, serving as a means to modulate intracellular metabolic fluxes. Recent strategies in this field have employed constraint-based dynamic models for process optimization, control, and estimation. However, this results in bilevel dynamic optimization problems, which pose considerable numerical and conceptual challenges. In this study, we present an alternative hybrid physics-informed dynamic modeling framework for metabolic cybergenetics, aimed at simplifying optimization, control, and estimation tasks. By utilizing machine-learning surrogates, our approach effectively embeds the physics of metabolic networks into the process rates of structurally simpler macro-kinetic models coupled with gene expression. These surrogates, informed by flux balance analysis, link the domains of manipulatable intracellular enzymes to metabolic exchange fluxes. This ensures that critical knowledge captured by the system's metabolic network is preserved. The resulting models can be integrated into metabolic cybergenetic schemes involving single-level optimizations. Additionally, the hybrid modeling approach maintains the number of system states at a necessary minimum, easing the burden of process monitoring and estimation. Our hybrid physics-informed metabolic cybergenetic framework is demonstrated using a computational case study on the optogenetically-assisted production of itaconate by $\textit{Escherichia coli}$.
comment: 25 pages, 10 figures, journal submission (reviewed/accepted version)
♻ ☆ Tikhonov Regularization for Stochastic Non-Smooth Convex Optimization in Hilbert Spaces
To solve non-smooth convex optimization problems with a noisy gradient input, we analyze the global behavior of subgradient-like flows under stochastic errors. The objective function is composite, being equal to the sum of two convex functions, one being differentiable and the other potentially non-smooth. We then use stochastic differential inclusions where the drift term is minus the subgradient of the objective function, and the diffusion term is either bounded or square-integrable. In this context, under Lipschitz's continuity of the differentiable term and a growth condition of the non-smooth term, our first main result shows almost sure weak convergence of the trajectory process towards a minimizer of the objective function. Then, using Tikhonov regularization with a properly tuned vanishing parameter, we can obtain almost sure strong convergence of the trajectory towards the minimum norm solution. We find an explicit tuning of this parameter when our objective function satisfies a local error-bound inequality. We also provide a comprehensive complexity analysis by establishing several new pointwise and ergodic convergence rates in expectation for the convex and strongly convex case.
comment: 34 pages, 2 tables. arXiv admin note: text overlap with arXiv:2207.02750
♻ ☆ Tight Convergence Rate Bounds for Optimization Under Power Law Spectral Conditions
Performance of optimization on quadratic problems sensitively depends on the low-lying part of the spectrum. For large (effectively infinite-dimensional) problems, this part of the spectrum can often be naturally represented or approximated by power law distributions, resulting in power law convergence rates for iterative solutions of these problems by gradient-based algorithms. In this paper, we propose a new spectral condition providing tighter upper bounds for problems with power law optimization trajectories. We use this condition to build a complete picture of upper and lower bounds for a wide range of optimization algorithms -- Gradient Descent, Steepest Descent, Heavy Ball, and Conjugate Gradients -- with an emphasis on the underlying schedules of learning rate and momentum. In particular, we demonstrate how an optimally accelerated method, its schedule, and convergence upper bound can be obtained in a unified manner for a given shape of the spectrum. Also, we provide first proofs of tight lower bounds for convergence rates of Steepest Descent and Conjugate Gradients under spectral power laws with general exponents. Our experiments show that the obtained convergence bounds and acceleration strategies are not only relevant for exactly quadratic optimization problems, but also fairly accurate when applied to the training of neural networks.
♻ ☆ Aerial Base Station Placement via Propagation Radio Maps
The deployment of aerial base stations (ABSs) on unmanned aerial vehicles (UAVs) presents a promising solution for extending cellular connectivity to areas where terrestrial infrastructure is overloaded, damaged, or absent. A pivotal challenge in this domain is to decide the locations of a set of ABSs to effectively serve ground-based users. Most existing approaches oversimplify this problem by assuming that the channel gain between two points is a function of solely distance and, sometimes, also the elevation angle. In turn, this paper leverages propagation radio maps to account for arbitrary air-to-ground channel gains. This methodology enables the identification of an approximately minimal set of locations where ABSs need to be deployed to ensure that all ground terminals achieve a target service rate, while adhering to backhaul capacity limitations and avoiding designated no-fly zones. Relying on a convex relaxation technique and the alternating direction method of multipliers (ADMM), this paper puts forth a scalable solver whose computational complexity scales linearly with the number of ground terminals. Convergence is established analytically and an extensive set of simulations corroborate the merits of the proposed scheme relative to conventional methods.
♻ ☆ Extremely Fast Convergence Rates for Extremum Seeking Control with Polyak-Ruppert Averaging
Stochastic approximation is a foundation for many algorithms found in machine learning and optimization. It is in general slow to converge: the mean square error vanishes as $O(n^{-1})$. A deterministic counterpart known as quasi-stochastic approximation is a viable alternative in many applications, including gradient-free optimization and reinforcement learning. It was assumed in prior research that the optimal achievable convergence rate is $O(n^{-2})$. It is shown in this paper that through design it is possible to obtain far faster convergence, of order $O(n^{-4+\delta})$, with $\delta>0$ arbitrary. Two techniques are introduced for the first time to achieve this rate of convergence. The theory is also specialized within the context of gradient-free optimization, and tested on standard benchmarks. The main results are based on a combination of novel application of results from number theory and techniques adapted from stochastic approximation theory.
comment: 36 pages, 14 figures
Systems and Control 65
☆ A Branch and Bound method for the exact parameter identification of the PK/PD model for anesthetic drugs
We address the problem of parameter identification for the standard pharmacokinetic/pharmacodynamic (PK/PD) model for anesthetic drugs. Our main contribution is the development of a global optimization method that guarantees finding the parameters that minimize the one-step ahead prediction error. The method is based on a branch-and-bound algorithm, that can be applied to solve a more general class of nonlinear regression problems. We present some simulation results, based on a dataset of twelve patients. In these simulations, we are always able to identify the exact parameters, despite the non-convexity of the overall identification problem.
☆ Looking back and forward: A retrospective and future directions on Software Engineering for systems-of-systems
Modern systems are increasingly connected and more integrated with other existing systems, giving rise to systems-of-systems (SoS). An SoS consists of a set of independent, heterogeneous systems that interact to provide new functionalities and accomplish global missions through emergent behavior manifested at runtime. The distinctive characteristics of SoS, when contrasted to traditional systems, pose significant research challenges within Software Engineering. These challenges motivate the need for a paradigm shift and the exploration of novel approaches for designing, developing, deploying, and evolving these systems. The International Workshop on Software Engineering for Systems-of-Systems (SESoS) series started in 2013 to fill a gap in scientific forums addressing SoS from the Software Engineering perspective, becoming the first venue for this purpose. This article presents a study aimed at outlining the evolution and future trajectory of Software Engineering for SoS based on the examination of 57 papers spanning the 11 editions of the SESoS workshop (2013-2023). The study combined scoping review and scientometric analysis methods to categorize and analyze the research contributions concerning temporal and geographic distribution, topics of interest, research methodologies employed, application domains, and research impact. Based on such a comprehensive overview, this article discusses current and future directions in Software Engineering for SoS.
☆ A data-based comparison of methods for reducing the peak volume flow rate in a district heating system
This work concerns reduction of the peak flow rate of a district heating grid, a key system property which is bounded by pipe dimensions and pumping capacity. The peak flow rate constrains the number of additional consumers that can be connected, and may be a limiting factor in reducing supply temperatures when transitioning to the 4th generation of district heating. We evaluate a full year of operational data from a subset of customer meters in a district heating system in Germany. We consider the peak flow rate reduction that could be achieved with full a posteriori knowledge of this data. Three strategies for reducing the peak flow rate are investigated: A load shifting demand response strategy, an upper limitation in substation return temperatures, and an upper limitation on each substation's volume flow rate. We show that imposing up to to 18 % load flexibility for the customers provides an equal reduction in the peak system flow rate under the load shifting strategy. The limited return temperature strategy is less efficient at curtailing the peak flow rate, but provides an overall reduction of volume flow rates. Finally, the flow rate limitation method can introduce new, higher flow rate peaks, reducing performance.
☆ SIS epidemics on open networks: A replacement-based approximation
In this paper we analyze continuous-time SIS epidemics subject to arrivals and departures of agents, by using an approximated process based on replacements. In defining the SIS dynamics in an open network, we consider a stochastic setting in which arrivals and departures take place according to Poisson processes with similar rates, and the new value of the infection probability of an arriving agent is drawn from a continuous distribution. Since the system size changes with time, we define an approximated process, in which replacements take place instead of arrivals and departures, and we focus on the evolution of an aggregate measure of the level of infection. So long as the reproduction number is less than one, the long-term behavior of this function measures the impact of the changes of the set of agents in the epidemic. We derive upper bounds for the expectation and variance of this function and we include a numerical example to show that the approximated process is close to the original SIS process.
comment: 7 pages, 2 figures, to appear in European Control Conference (ECC 2024)
☆ Predictable Interval MDPs through Entropy Regularization
Regularization of control policies using entropy can be instrumental in adjusting predictability of real-world systems. Applications benefiting from such approaches range from, e.g., cybersecurity, which aims at maximal unpredictability, to human-robot interaction, where predictable behavior is highly desirable. In this paper, we consider entropy regularization for interval Markov decision processes (IMDPs). IMDPs are uncertain MDPs, where transition probabilities are only known to belong to intervals. Lately, IMDPs have gained significant popularity in the context of abstracting stochastic systems for control design. In this work, we address robust minimization of the linear combination of entropy and a standard cumulative cost in IMDPs, thereby establishing a trade-off between optimality and predictability. We show that optimal deterministic policies exist, and devise a value-iteration algorithm to compute them. The algorithm solves a number of convex programs at each step. Finally, through an illustrative example we show the benefits of penalizing entropy in IMDPs.
☆ BatDeck: Advancing Nano-drone Navigation with Low-power Ultrasound-based Obstacle Avoidance
Nano-drones, distinguished by their agility, minimal weight, and cost-effectiveness, are particularly well-suited for exploration in confined, cluttered and narrow spaces. Recognizing transparent, highly reflective or absorbing materials, such as glass and metallic surfaces is challenging, as classical sensors, such as cameras or laser rangers, often do not detect them. Inspired by bats, which can fly at high speeds in complete darkness with the help of ultrasound, this paper introduces \textit{BatDeck}, a pioneering sensor-deck employing a lightweight and low-power ultrasonic sensor for nano-drone autonomous navigation. This paper first provides insights about sensor characteristics, highlighting the influence of motor noise on the ultrasound readings, then it introduces the results of extensive experimental tests for obstacle avoidance (OA) in a diverse environment. Results show that \textit{BatDeck} allows exploration for a flight time of 8 minutes while covering 136m on average before crash in a challenging environment with transparent and reflective obstacles, proving the effectiveness of ultrasonic sensors for OA on nano-drones.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
Trajectory Planning of Robotic Manipulator in Dynamic Environment Exploiting DRL
This study is about the implementation of a reinforcement learning algorithm in the trajectory planning of manipulators. We have a 7-DOF robotic arm to pick and place the randomly placed block at a random target point in an unknown environment. The obstacle is randomly moving which creates a hurdle in picking the object. The objective of the robot is to avoid the obstacle and pick the block with constraints to a fixed timestamp. In this literature, we have applied a deep deterministic policy gradient (DDPG) algorithm and compared the model's efficiency with dense and sparse rewards.
comment: Accepted in ICIESTR-2024
☆ Symbolic and User-friendly Geometric Algebra Routines (SUGAR) for Computations in Matlab
Geometric algebra (GA) is a mathematical tool for geometric computing, providing a framework that allows a unified and compact approach to geometric relations which in other mathematical systems are typically described using different more complicated elements. This fact has led to an increasing adoption of GA in applied mathematics and engineering problems. However, the scarcity of symbolic implementations of GA and its inherent complexity, requiring a specific mathematical background, make it challenging and less intuitive for engineers to work with. This prevents wider adoption among more applied professionals. To address this challenge, this paper introduces SUGAR (Symbolic and User-friendly Geometric Algebra Routines), an open-source toolbox designed for Matlab and licensed under the MIT License. SUGAR facilitates the translation of GA concepts into Matlab and provides a collection of user-friendly functions tailored for GA computations, including support for symbolic operations. It supports both numeric and symbolic computations in high-dimensional GAs. Specifically tailored for applied mathematics and engineering applications, SUGAR has been meticulously engineered to represent geometric elements and transformations within two and three-dimensional projective and conformal geometric algebras, aligning with established computational methodologies in the literature. Furthermore, SUGAR efficiently handles functions of multivectors, such as exponential, logarithmic, sinusoidal, and cosine functions, enhancing its applicability across various engineering domains, including robotics, control systems, and power electronics. Finally, this work includes four distinct validation examples, demonstrating SUGAR's capabilities across the above-mentioned fields and its practical utility in addressing real-world applied mathematics and engineering problems.
comment: 33 pages, 6 figures, journal paper submitted to ACM TOMS
☆ Guided Bayesian Optimization: Data-Efficient Controller Tuning with Digital Twin
This article presents the guided Bayesian optimization algorithm as an efficient data-driven method for iteratively tuning closed-loop controller parameters using an event-triggered digital twin of the system based on available closed-loop data. We define a controller tuning framework independent of the controller or the plant structure. Our proposed methodology is model-free, making it suitable for nonlinear and unmodelled plants with measurement noise. The objective function consists of performance metrics modeled by Gaussian processes. We utilize the available information in the closed-loop system to identify and progressively maintain a digital twin that guides the optimizer, improving the data efficiency of our method. Switching the digital twin on and off is triggered by data-driven criteria related to the digital twin's uncertainty estimations in the BO tuning framework. Effectively, it replaces much of the exploration of the real system with exploration performed on the digital twin. We analyze the properties of our method in simulation and demonstrate its performance on two real closed-loop systems with different plant and controller structures. The experimental results show that our method requires fewer experiments on the physical plant than Bayesian optimization to find the optimal controller parameters.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
☆ The Adaptive Workplace: Orchestrating Architectural Services around the Wellbeing of Individual Occupants
As the academic consortia members of the EU Horizon project SONATA ("Situation-aware OrchestratioN of AdapTive Architecture"), we respond to the workshop call for "Office Wellbeing by Design: Don't Stand for Anything Less" by proposing the "Adaptive Workplace" concept. In essence, our vision aims to adapt a workplace to the ever-changing needs of individual occupants, instead of that occupants are expected to adapt to their workplace.
☆ Counter-example guided Imitation Learning of Feedback Controllers from Temporal Logic Specifications
We present a novel method for imitation learning for control requirements expressed using Signal Temporal Logic (STL). More concretely we focus on the problem of training a neural network to imitate a complex controller. The learning process is guided by efficient data aggregation based on counter-examples and a coverage measure. Moreover, we introduce a method to evaluate the performance of the learned controller via parameterization and parameter estimation of the STL requirements. We demonstrate our approach with a flying robot case study.
☆ Sparsity-Constrained Linear Quadratic Regulation Problem: Greedy Approach with Performance Guarantee
We study a linear quadratic regulation problem with a constraint where the control input can be nonzero only at a limited number of times. Given that this constraint leads to a combinational optimization problem, we adopt a greedy method to find a suboptimal solution. To quantify the performance of the greedy algorithm, we employ two metrics that reflect the submodularity level of the objective function: The submodularity ratio and curvature. We first present an explicit form of the optimal control input that is amenable to evaluating these metrics. Subsequently, we establish bounds on the submodularity ratio and curvature, which enable us to offer a practical performance guarantee for the greedy algorithm. The effectiveness of our guarantee is further demonstrated through numerical simulations.
comment: 8 pages, 3 figures
☆ Decoupling parameter variation from noise: Biquadratic Lyapunov forms in data-driven LPV control
A promising step from linear towards nonlinear data-driven control is via the design of controllers for linear parameter-varying (LPV) systems, which are linear systems whose parameters are varying along a measurable scheduling signal. However, the interplay between uncertainty arising from corrupted data and the parameter-varying nature of these systems impacts the stability analysis, and limits the generalization of well-understood data-driven methods for linear time-invariant systems. In this work, we decouple this interplay using a recently developed variant of the Fundamental Lemma for LPV systems and the viewpoint of data-informativity, in combination with biquadratic Lyapunov forms. Together, these allow us to develop novel linear matrix inequality conditions for the existence of scheduling-dependent Lyapunov functions, incorporating the intrinsic nonlinearity. Appealingly, these results are stated purely in terms of the collected data and bounds on the noise, and they are computationally favorable to check.
comment: Submitted for CDC 2024
☆ Hybrid low-dimensional limiting state of charge estimator for multi-cell lithium-ion batteries
The state of charge (SOC) of lithium-ion batteries needs to be accurately estimated for safety and reliability purposes. For battery packs made of a large number of cells, it is not always feasible to design one SOC estimator per cell due to limited computational resources. Instead, only the minimum and the maximum SOC need to be estimated. The challenge is that the cells having minimum and maximum SOC typically change over time. In this context, we present a low-dimensional hybrid estimator of the minimum (maximum) SOC, whose convergence is analytically guaranteed. We consider for this purpose a battery consisting of cells interconnected in series, which we model by electric equivalent circuit models. We then present the hybrid estimator, which runs an observer designed for a single cell at any time instant, selected by a switching-like logic mechanism. We establish a practical exponential stability property for the estimation error on the minimum (maximum) SOC thereby guaranteeing the ability of the hybrid scheme to generate accurate estimates of the minimum (maximum) SOC. The analysis relies on non-smooth hybrid Lyapunov techniques. A numerical illustration is provided to showcase the relevance of the proposed approach.
☆ Spatially temporally distributed informative path planning for multi-robot systems
This paper investigates the problem of informative path planning for a mobile robotic sensor network in spatially temporally distributed mapping. The robots are able to gather noisy measurements from an area of interest during their movements to build a Gaussian Process (GP) model of a spatio-temporal field. The model is then utilized to predict the spatio-temporal phenomenon at different points of interest. To spatially and temporally navigate the group of robots so that they can optimally acquire maximal information gains while their connectivity is preserved, we propose a novel multistep prediction informative path planning optimization strategy employing our newly defined local cost functions. By using the dual decomposition method, it is feasible and practical to effectively solve the optimization problem in a distributed manner. The proposed method was validated through synthetic experiments utilizing real-world data sets.
☆ Ensuring Disturbance Rejection Performance by Synthesizing Grid-Following and Grid-Forming Inverters in Power Systems
To meet the dynamic requirement of power systems, it is imperative for grid-connected inverters to ensure good disturbance rejection performance (DRP) under variable grid conditions. This letter discovers and theoretically proves that for the general networks, synthesizing grid-following (GFL) inverters and grid-forming (GFM) inverters can more effectively ensure the DRP of multiple inverters, compared to homogeneous inverter-based systems that solely use either GFL or GFM inverters. The combination of GFL inverters and GFM inverters can concurrently increases the smallest eigenvalue and decreases the largest eigenvalue of the grounded network Laplacian matrix. This can be equivalent to rematching the proper short-circuit ratio (SCR) for GFL and GFM inverters, thereby ensuring the DRP of inverters both in weak and strong grids. The result reveals the necessity of synthesizing diverse inverter control schemes from the network-based perspective. Sensitivity function-based analysis and real-time simulations confirm the effectiveness of our results.
comment: 5 pages
☆ Data-Driven Extrusion Force Control Tuning for 3D Printing
The quality of 3D prints often varies due to different conditions inherent to each print, such as filament type, print speed, and nozzle size. Closed-loop process control methods improve the accuracy and repeatability of 3D prints. However, optimal tuning of controllers for given process parameters and design geometry is often a challenge with manually tuned controllers resulting in inconsistent and suboptimal results. This work employs Bayesian optimization to identify the optimal controller parameters. Additionally, we explore transfer learning in the context of 3D printing by leveraging prior information from past trials. By integrating optimized extrusion force control and transfer learning, we provide a novel framework for closed-loop 3D printing and propose an automated calibration routine that produces high-quality prints for a desired combination of print settings, material, and shape.
comment: Submitted to IEEE CASE 2024
☆ A Geometric Perspective on Fusing Gaussian Distributions on Lie Groups
Stochastic inference on Lie groups plays a key role in state estimation problems, such as inertial navigation, visual inertial odometry, pose estimation in virtual reality, etc. A key problem is fusing independent concentrated Gaussian distributions defined at different reference points on the group. In this paper we approximate distributions at different points in the group in a single set of exponential coordinates and then use classical Gaussian fusion to obtain the fused posteriori in those coordinates. We consider several approximations including the exact Jacobian of the change of coordinate map, first and second order Taylor's expansions of the Jacobian, and parallel transport with and without curvature correction associated with the underlying geometry of the Lie group. Preliminary results on SO(3) demonstrate that a novel approximation using parallel transport with curvature correction achieves similar accuracy to the state-of-the-art optimisation based algorithms at a fraction of the computational cost.
comment: Preprint for L-CSS
☆ A Distributionally Robust Model Predictive Control for Static and Dynamic Uncertainties in Smart Grids
The integration of various power sources, including renewables and electric vehicles, into smart grids is expanding, introducing uncertainties that can result in issues like voltage imbalances, load fluctuations, and power losses. These challenges negatively impact the reliability and stability of online scheduling in smart grids. Existing research often addresses uncertainties affecting current states but overlooks those that impact future states, such as the unpredictable charging patterns of electric vehicles. To distinguish between these, we term them static uncertainties and dynamic uncertainties, respectively. This paper introduces WDR-MPC, a novel approach that stands for two-stage Wasserstein-based Distributionally Robust (WDR) optimization within a Model Predictive Control (MPC) framework, aimed at effectively managing both types of uncertainties in smart grids. The dynamic uncertainties are first reformulated into ambiguity tubes and then the distributionally robust bounds of both dynamic and static uncertainties can be established using WDR optimization. By employing ambiguity tubes and WDR optimization, the stochastic MPC system is converted into a nominal one. Moreover, we develop a convex reformulation method to speed up WDR computation during the two-stage optimization. The distinctive contribution of this paper lies in its holistic approach to both static and dynamic uncertainties in smart grids. Comprehensive experiment results on IEEE 38-bus and 94-bus systems reveal the method's superior performance and the potential to enhance grid stability and reliability.
☆ Physics-informed RL for Maximal Safety Probability Estimation
Accurate risk quantification and reachability analysis are crucial for safe control and learning, but sampling from rare events, risky states, or long-term trajectories can be prohibitively costly. Motivated by this, we study how to estimate the long-term safety probability of maximally safe actions without sufficient coverage of samples from risky states and long-term trajectories. The use of maximal safety probability in control and learning is expected to avoid conservative behaviors due to over-approximation of risk. Here, we first show that long-term safety probability, which is multiplicative in time, can be converted into additive costs and be solved using standard reinforcement learning methods. We then derive this probability as solutions of partial differential equations (PDEs) and propose Physics-Informed Reinforcement Learning (PIRL) algorithm. The proposed method can learn using sparse rewards because the physics constraints help propagate risk information through neighbors. This suggests that, for the purpose of extracting more information for efficient learning, physics constraints can serve as an alternative to reward shaping. The proposed method can also estimate long-term risk using short-term samples and deduce the risk of unsampled states. This feature is in stark contrast with the unconstrained deep RL that demands sufficient data coverage. These merits of the proposed method are demonstrated in numerical simulation.
Optimal Operation of Reconfigurable Active Distribution Networks Aiming at Resiliency Improvement
As natural disasters bring about power outage and financial losses, network resiliency is an important challenge for distribution network operators (DNOs). On the other side, power loss reduction during normal operating condition is a major concern of DNOs. In this paper, optimal scheduling of active distribution network (ADN) is addressed through simultaneous minimization of power loss in normal condition and load shedding in critical condition after natural disasters. A new formulation is developed for the network reconfiguration to optimize the system operation in both normal and emergency conditions in the presence of conventional and renewable-energy-based distributed generation (DG) as well as energy storage systems (ESSs). The line flow based (LFB) algorithm is used for the AC power flow calculations, and all the developed relations have been convexified to construct a mixed-integer quadratically-constrained programming (MIQCP) optimization model. The simulations have been implemented on the IEEE 33-bus system in GAMS, and the results are investigated.
☆ Impact-Aware Bimanual Catching of Large-Momentum Objects
This paper investigates one of the most challenging tasks in dynamic manipulation -- catching large-momentum moving objects. Beyond the realm of quasi-static manipulation, dealing with highly dynamic objects can significantly improve the robot's capability of interacting with its surrounding environment. Yet, the inevitable motion mismatch between the fast moving object and the approaching robot will result in large impulsive forces, which lead to the unstable contacts and irreversible damage to both the object and the robot. To address the above problems, we propose an online optimization framework to: 1) estimate and predict the linear and angular motion of the object; 2) search and select the optimal contact locations across every surface of the object to mitigate impact through sequential quadratic programming (SQP); 3) simultaneously optimize the end-effector motion, stiffness, and contact force for both robots using multi-mode trajectory optimization (MMTO); and 4) realise the impact-aware catching motion on the compliant robotic system based on indirect force controller. We validate the impulse distribution, contact selection, and impact-aware MMTO algorithms in simulation and demonstrate the benefits of the proposed framework in real-world experiments including catching large-momentum moving objects with well-defined motion, constrained motion and free-flying motion.
☆ DASA: Delay-Adaptive Multi-Agent Stochastic Approximation
We consider a setting in which $N$ agents aim to speedup a common Stochastic Approximation (SA) problem by acting in parallel and communicating with a central server. We assume that the up-link transmissions to the server are subject to asynchronous and potentially unbounded time-varying delays. To mitigate the effect of delays and stragglers while reaping the benefits of distributed computation, we propose \texttt{DASA}, a Delay-Adaptive algorithm for multi-agent Stochastic Approximation. We provide a finite-time analysis of \texttt{DASA} assuming that the agents' stochastic observation processes are independent Markov chains. Significantly advancing existing results, \texttt{DASA} is the first algorithm whose convergence rate depends only on the mixing time $\tmix$ and on the average delay $\tau_{avg}$ while jointly achieving an $N$-fold convergence speedup under Markovian sampling. Our work is relevant for various SA applications, including multi-agent and distributed temporal difference (TD) learning, Q-learning and stochastic optimization with correlated data.
☆ A Discrete-Time Least-Squares Adaptive State Tracking Control Scheme with A Mobile-Robot System Study
This paper develops an adaptive state tracking control scheme for discrete-time systems, using the least-squares algorithm, as the new solution to the long-standing discrete-time adaptive state tracking control problem to which the Lyapunov method (well-developed for the continuous-time adaptive state tracking problem) is not applicable. The new adaptive state tracking scheme is based on a recently-developed new discrete-time error model which has been used for gradient algorithm based state tracking control schemes, and uses the least-squares algorithm for parameter adaptation. The new least-squares algorithm is derived to minimize an accumulative estimation error, to ensure certain optimality for parameter estimation. The system stability and output tracking properties are studied. Technical results are presented in terms of plant-model matching, error model, adaptive law, optimality formulation, and stability and tracking analysis. The developed adaptive control scheme is applied to a discrete-time multiple mobile robot system to meet an adaptive state tracking objective. In addition, a collision avoidance mechanism is proposed to prevent collisions in the whole tracking process. Simulation results are presented, which verify the desired system state tracking properties under the developed least-squares algorithm based adaptive control scheme.
☆ Active Learning of Dynamics Using Prior Domain Knowledge in the Sampling Process
We present an active learning algorithm for learning dynamics that leverages side information by explicitly incorporating prior domain knowledge into the sampling process. Our proposed algorithm guides the exploration toward regions that demonstrate high empirical discrepancy between the observed data and an imperfect prior model of the dynamics derived from side information. Through numerical experiments, we demonstrate that this strategy explores regions of high discrepancy and accelerates learning while simultaneously reducing model uncertainty. We rigorously prove that our active learning algorithm yields a consistent estimate of the underlying dynamics by providing an explicit rate of convergence for the maximum predictive variance. We demonstrate the efficacy of our approach on an under-actuated pendulum system and on the half-cheetah MuJoCo environment.
☆ High-dimensional continuification control of large-scale multi-agent systems under limited sensing and perturbations
This paper investigates the robustness of a novel high-dimensional continuification control method for complex multi-agent systems. We begin by formulating a partial differential equation describing the spatio-temporal density dynamics of swarming agents. A stable control action for the density is then derived and validated under nominal conditions. Subsequently, we discretize this macroscopic strategy into actionable velocity inputs for the system's agents. Our analysis demonstrates the robustness of the approach beyond idealized assumptions of unlimited sensing and absence of perturbations.
comment: arXiv admin note: text overlap with arXiv:2310.01573
☆ Robust Finite-time Stabilization of Linear Systems with Limited State Quantization
This paper investigates the robust asymptotic stabilization of a linear time-invariant (LTI) system by a static feedback with a static state quantization. It is shown that the controllable LTI system can be stabilized to zero in a finite time by means of a nonlinear feedback with a quantizer having a limited (finite) number of values (quantization seeds) even when all parameters of the controller and the quantizer are time-invariant. The control design is based on generalized homogeneity. A homogeneous spherical quantizer is introduced. The static homogeneous feedback is shown to be local (or global) finite-time stabilizer for the linear system (dependently of the system matrix). The tuning rules for both the quantizer and the feedback law are obtained in the form of Linear Matrix Inequalities (LMIs). The closed-loop system is proven to be robust with respect to some bounded matched and vanishing mismatched perturbations. Theoretical results are supported by numerical simulations. \
☆ Belief Samples Are All You Need For Social Learning
In this paper, we consider the problem of social learning, where a group of agents embedded in a social network are interested in learning an underlying state of the world. Agents have incomplete, noisy, and heterogeneous sources of information, providing them with recurring private observations of the underlying state of the world. Agents can share their learning experience with their peers by taking actions observable to them, with values from a finite feasible set of states. Actions can be interpreted as samples from the beliefs which agents may form and update on what the true state of the world is. Sharing samples, in place of full beliefs, is motivated by the limited communication, cognitive, and information-processing resources available to agents especially in large populations. Previous work (Salhab et al.) poses the question as to whether learning with probability one is still achievable if agents are only allowed to communicate samples from their beliefs. We provide a definite positive answer to this question, assuming a strongly connected network and a ``collective distinguishability'' assumption, which are both required for learning even in full-belief-sharing settings. In our proposed belief update mechanism, each agent's belief is a normalized weighted geometric interpolation between a fully Bayesian private belief -- aggregating information from the private source -- and an ensemble of empirical distributions of the samples shared by her neighbors over time. By carefully constructing asymptotic almost-sure lower/upper bounds on the frequency of shared samples matching the true state/or not, we rigorously prove the convergence of all the beliefs to the true state, with probability one.
comment: 6 pages
☆ Output-feedback Synthesis Orbit Geometry: Quotient Manifolds and LQG Direct Policy Optimization
In this paper, we consider direct policy optimization for the linear-quadratic Gaussian (LQG) setting. Over the past few years, it has been recognized that the landscape of stabilizing output-feedback controllers of relevance to LQG has an intricate geometry, particularly as it pertains to the existence of spurious stationary points. In order to address such challenges, in this paper, we first adopt a Riemannian metric for the space of stabilizing full-order minimal output-feedback controllers. We then proceed to prove that the orbit of such controllers modulo coordinate transformation admits a Riemannian quotient manifold structure. This geometric structure is then used to develop a Riemannian gradient descent for the direct LQG policy optimization. We prove a local convergence guarantee with linear rate and show the proposed approach exhibits significantly faster and more robust numerical performance as compared with ordinary gradient descent for LQG. Subsequently, we provide reasons for this observed behavior; in particular, we argue that optimizing over the orbit space of controllers is the right theoretical and computational setup for direct LQG policy optimization.
☆ Approximation with Random Shallow ReLU Networks with Applications to Model Reference Adaptive Control
Neural networks are regularly employed in adaptive control of nonlinear systems and related methods o reinforcement learning. A common architecture uses a neural network with a single hidden layer (i.e. a shallow network), in which the weights and biases are fixed in advance and only the output layer is trained. While classical results show that there exist neural networks of this type that can approximate arbitrary continuous functions over bounded regions, they are non-constructive, and the networks used in practice have no approximation guarantees. Thus, the approximation properties required for control with neural networks are assumed, rather than proved. In this paper, we aim to fill this gap by showing that for sufficiently smooth functions, ReLU networks with randomly generated weights and biases achieve $L_{\infty}$ error of $O(m^{-1/2})$ with high probability, where $m$ is the number of neurons. It suffices to generate the weights uniformly over a sphere and the biases uniformly over an interval. We show how the result can be used to get approximations of required accuracy in a model reference adaptive control application.
comment: Under Review for Conference on Decision and Control
☆ Adaptive Step Duration for Precise Foot Placement: Achieving Robust Bipedal Locomotion on Terrains with Restricted Footholds
This paper introduces a novel multi-step preview foot placement planning algorithm designed to enhance the robustness of bipedal robotic walking across challenging terrains with restricted footholds. Traditional one-step preview planning struggles to maintain stability when stepping areas are severely limited, such as with random stepping stones. In this work, we developed a discrete-time Model Predictive Control (MPC) based on the step-to-step discrete evolution of the Divergent Component of Motion (DCM) of bipedal locomotion. This approach adaptively changes the step duration for optimal foot placement under constraints, thereby ensuring the robot's operational viability over multiple future steps and significantly improving its ability to navigate through environments with tight constraints on possible footholds. The effectiveness of this planning algorithm is demonstrated through simulations that include a variety of complex stepping-stone configurations and external perturbations. These tests underscore the algorithm's improved performance for navigating foothold-restricted environments, even with the presence of external disturbances.
comment: 8 pages, 8 figures, submitted to CDC 2024, for associated simulation video, see https://youtu.be/2jhikPlZmbE
☆ State-Augmented Linear Games with Antagonistic Error for High-Dimensional, Nonlinear Hamilton-Jacobi Reachability
Hamilton-Jacobi Reachability (HJR) is a popular method for analyzing the liveness and safety of a dynamical system with bounded control and disturbance. The corresponding HJ value function offers a robust controller and characterizes the reachable sets, but is traditionally solved with Dynamic Programming (DP) and limited to systems of dimension less than six. Recently, the space-parallelizeable, generalized Hopf formula has been shown to also solve the HJ value with a nearly three-log increase in dimension limit, but is limited to linear systems. To extend this potential, we demonstrate how state-augmented (SA) spaces, which are well-known for their improved linearization accuracy, may be used to solve tighter, conservative approximations of the value function with any linear model in this SA space. Namely, we show that with a representation of the true dynamics in the SA space, a series of inequalities confirms that the value of a SA linear game with antagonistic error is a conservative envelope of the true value function. It follows that if the optimal controller for the HJ SA linear game with error may succeed, it will also succeed in the true system. Unlike previous methods, this result offers the ability to safely approximate reachable sets and their corresponding controllers with the Hopf formula in a non-convex manner. Finally, we demonstrate this in the slow manifold system for clarity, and in the controlled Van der Pol system with different lifting functions.
☆ An Optimal Solution to Infinite Horizon Nonlinear Control Problems: Part II
This paper considers the infinite horizon optimal control problem for nonlinear systems. Under the condition of nonlinear controllability of the system to any terminal set containing the origin and forward invariance of the terminal set, we establish a regularized solution approach consisting of a ``finite free final time" optimal transfer problem to the terminal set which renders the set globally asymptotically stable. Further, we show that the approximations converge to the optimal infinite horizon cost as the size of the terminal set decreases to zero. We also perform the analysis for the discounted problem and show that the terminal set is asymptotically stable only for a subset of the state space and not globally. The theory is empirically evaluated on various nonholonomic robotic systems to show that the cost of our approximate problem converges and the transfer time into the terminal set is dependent on the initial state of the system, necessitating the free final time formulation.
☆ Bayesian Methods for Trust in Collaborative Multi-Agent Autonomy
Multi-agent, collaborative sensor fusion is a vital component of a multi-national intelligence toolkit. In safety-critical and/or contested environments, adversaries may infiltrate and compromise a number of agents. We analyze state of the art multi-target tracking algorithms under this compromised agent threat model. We prove that the track existence probability test ("track score") is significantly vulnerable to even small numbers of adversaries. To add security awareness, we design a trust estimation framework using hierarchical Bayesian updating. Our framework builds beliefs of trust on tracks and agents by mapping sensor measurements to trust pseudomeasurements (PSMs) and incorporating prior trust beliefs in a Bayesian context. In case studies, our trust estimation algorithm accurately estimates the trustworthiness of tracks/agents, subject to observability limitations.
☆ Spline Trajectory Tracking and Obstacle Avoidance for Mobile Agents via Convex Optimization
We propose an output feedback control-based motion planning technique for agents to enable them to converge to a specified polynomial trajectory while imposing a set of safety constraints on our controller to avoid collisions within the free configuration space (polygonal environment). To achieve this, we 1) decompose our polygonal environment into different overlapping cells 2) write out our polynomial trajectories as the output of a reference dynamical system with given initial conditions 3) formulate convergence and safety constraints as Linear Matrix Inequalities (LMIs) on our controller using Control Lyapunov Functions (CLFs) and Control Barrier Functions (CBFs) and 4) solve a semi-definite programming (SDP) problem with convergence and safety constraints imposed to synthesize a controller for each convex cell. Extensive simulations are included to test our motion planning method under different initial conditions and different reference trajectories. The synthesized controller is robust to changes in initial conditions and is always safe relative to the boundaries of the polygonal environment.
☆ State Space Models as Foundation Models: A Control Theoretic Overview
In recent years, there has been a growing interest in integrating linear state-space models (SSM) in deep neural network architectures of foundation models. This is exemplified by the recent success of Mamba, showing better performance than the state-of-the-art Transformer architectures in language tasks. Foundation models, like e.g. GPT-4, aim to encode sequential data into a latent space in order to learn a compressed representation of the data. The same goal has been pursued by control theorists using SSMs to efficiently model dynamical systems. Therefore, SSMs can be naturally connected to deep sequence modeling, offering the opportunity to create synergies between the corresponding research areas. This paper is intended as a gentle introduction to SSM-based architectures for control theorists and summarizes the latest research developments. It provides a systematic review of the most successful SSM proposals and highlights their main features from a control theoretic perspective. Additionally, we present a comparative analysis of these models, evaluating their performance on a standardized benchmark designed for assessing a model's efficiency at learning long sequences.
☆ A Semi-Lagrangian Approach for Time and Energy Path Planning Optimization in Static Flow Fields
Efficient path planning for autonomous mobile robots is a critical problem across numerous domains, where optimizing both time and energy consumption is paramount. This paper introduces a novel methodology that considers the dynamic influence of an environmental flow field and considers geometric constraints, including obstacles and forbidden zones, enriching the complexity of the planning problem. We formulate it as a multi-objective optimal control problem, propose a novel transformation called Harmonic Transformation, and apply a semi-Lagrangian scheme to solve it. The set of Pareto efficient solutions is obtained considering two distinct approaches: a deterministic method and an evolutionary-based one, both of which are designed to make use of the proposed Harmonic Transformation. Through an extensive analysis of these approaches, we demonstrate their efficacy in finding optimized paths.
comment: 12 pages, initial paper submission; Preprint submitted to the IEEE Transactions on Intelligent Transportation Systems
☆ Semantic-Aware Remote Estimation of Multiple Markov Sources Under Constraints
This paper studies semantic-aware communication for remote estimation of multiple Markov sources over a lossy and rate-constrained channel. Unlike most existing studies that treat all source states equally, we exploit the semantics of information and consider that the remote actuator has different tolerances for the estimation errors of different states. We aim to find an optimal scheduling policy that minimizes the long-term state-dependent costs of estimation errors under a transmission frequency constraint. We theoretically show the structure of the optimal policy by leveraging the average-cost Constrained Markov Decision Process (CMDP) theory and the Lagrangian dynamic programming. By exploiting the optimal structural results, we develop a novel policy search algorithm, termed intersection search plus relative value iteration (Insec-RVI), that can find the optimal policy using only a few iterations. To avoid the ``curse of dimensionality'' of MDPs, we propose an online low-complexity drift-plus-penalty (DPP) scheduling algorithm based on the Lyapunov optimization theorem. We also design an efficient average-cost Q-learning algorithm to estimate the optimal policy without knowing a priori the channel and source statistics. Numerical results show that continuous transmission is inefficient, and remarkably, our semantic-aware policies can attain the optimum by strategically utilizing fewer transmissions by exploiting the timing of the important information.
☆ Energy Efficiency Optimization Method of WDM Visible Light Communication System for Indoor Broadcasting Networks
This paper introduces a novel approach to optimize energy efficiency in wavelength division multiplexing (WDM) Visible Light Communication (VLC) systems designed for indoor broadcasting networks. A physics-based LED model is integrated into system energy efficiency optimization, enabling quantitative analysis of the critical issue of VLC energy efficiency: the nonlinear interplay between illumination and communication performance. The optimization jointly incorporates constraints on communication quality of each channel, and illumination performance, standardized by the International Commission on Illumination (CIE). The formulated nonlinear optimization problem is solved by the Sequential Quadratic Programming (SQP) algorithm in an experiment-based simulation. An integrated Red-Green-Blue-Yellow Light Emitting Diode (RGBY-LED) is measured for model calibration and three different scenarios are simulated to evaluate the generality of the proposed method. Results demonstrate a double enhancement in performance and a high versatility in accommodating various scenarios. Furthermore, it highlights the importance of balancing communication and illumination imperatives in VLC systems, challenging conventional perceptions focused solely on minimizing power consumption.
☆ Resource and Mobility Management in Hybrid LiFi and WiFi Networks: A User-Centric Learning Approach
Hybrid light fidelity (LiFi) and wireless fidelity (WiFi) networks (HLWNets) are an emerging indoor wireless communication paradigm, which combines the advantages of the capacious optical spectra of LiFi and ubiquitous coverage of WiFi. Meanwhile, load balancing (LB) becomes a key challenge in resource management for such hybrid networks. The existing LB methods are mostly network-centric, relying on a central unit to make a solution for the users all at once. Consequently, the solution needs to be updated for all users at the same pace, regardless of their moving status. This would affect the network performance in two aspects: i) when the update frequency is low, it would compromise the connectivity of fast-moving users; ii) when the update frequency is high, it would cause unnecessary handovers as well as hefty feedback costs for slow-moving users. Motivated by this, we investigate user-centric LB which allows users to update their solutions at different paces. The research is developed upon our previous work on adaptive target-condition neural network (ATCNN), which can conduct LB for individual users in quasi-static channels. In this paper, a deep neural network (DNN) model is designed to enable an adaptive update interval for each individual user. This new model is termed as mobility-supporting neural network (MSNN). Associating MSNN with ATCNN, a user-centric LB framework named mobility-supporting ATCNN (MS-ATCNN) is proposed to handle resource management and mobility management simultaneously. Results show that at the same level of average update interval, MS-ATCNN can achieve a network throughput up to 215\% higher than conventional LB methods such as game theory, especially for a larger number of users. In addition, MS-ATCNN costs an ultra low runtime at the level of 100s $\mu$s, which is two to three orders of magnitude lower than game theory.
comment: 12 pages, 12 figures, 3 tables, submitted to IEEE TWC
☆ Scheduling Power-Intensive Operations of Battery Energy Storage Systems and Application to Hybrid Hydropower Plants
Classical schedulers for Battery Energy Storage Systems (BESSs) use static power constraints, assuming that the BESS can provide the rated power at any State-Of-Charge (SOC) level and that these are representative of the underlying physical constraints of the system (BESS voltage and current). Static power constraints, however, can generate unfeasible schedules, especially in power-intensive applications, as demonstrated in this paper. This paper derives a set of alternative constraints for the BESS power that are cognizant of the physical limits of the BESS. It is shown that these constraints, developed by leveraging an equivalent circuit model of the BESS, can be formulated as linear inequalities in scheduling problems, thus leaving the properties of the original problem (i.e., linearity and convexity) unaltered. A comparative analysis against traditional schedulers from the literature shows significant reductions in current violations and the generation of feasible schedules. These findings underscore the crucial role of implementing more advanced power constraints of BESSs in power-intensive applications, thereby enhancing the reliability of BESS scheduling strategies.
☆ An LLM-Based Digital Twin for Optimizing Human-in-the Loop Systems
The increasing prevalence of Cyber-Physical Systems and the Internet of Things (CPS-IoT) applications and Foundation Models are enabling new applications that leverage real-time control of the environment. For example, real-time control of Heating, Ventilation and Air-Conditioning (HVAC) systems can reduce its usage when not needed for the comfort of human occupants, hence reducing energy consumption. Collecting real-time feedback on human preferences in such human-in-the-loop (HITL) systems, however, is difficult in practice. We propose the use of large language models (LLMs) to deal with the challenges of dynamic environments and difficult-to-obtain data in CPS optimization. In this paper, we present a case study that employs LLM agents to mimic the behaviors and thermal preferences of various population groups (e.g. young families, the elderly) in a shopping mall. The aggregated thermal preferences are integrated into an agent-in-the-loop based reinforcement learning algorithm AitL-RL, which employs the LLM as a dynamic simulation of the physical environment to learn how to balance between energy savings and occupant comfort. Our results show that LLMs are capable of simulating complex population movements within large open spaces. Besides, AitL-RL demonstrates superior performance compared to the popular existing policy of set point control, suggesting that adaptive and personalized decision-making is critical for efficient optimization in CPS-IoT applications. Through this case study, we demonstrate the potential of integrating advanced Foundation Models like LLMs into CPS-IoT to enhance system adaptability and efficiency. The project's code can be found on our GitHub repository.
comment: Accepted at International Workshop on Foundation Models for Cyber-Physical Systems & Internet of Things (FMSys) 2024, Co-located at CPS-IoT Week 2024
☆ Privacy Preservation by Intermittent Transmission in Cooperative LQG Control Systems
In this paper, we study a cooperative linear quadratic Gaussian (LQG) control system with a single user and a server. In this system, the user runs a process and employs the server to meet the needs of computation. However, the user regards its state trajectories as privacy. Therefore, we propose a privacy scheme, in which the user sends data to the server intermittently. By this scheme, the server's received information of the user is reduced, and consequently the user's privacy is preserved. In this paper, we consider a periodic transmission scheme. We analyze the performance of privacy preservation and LQG control of different transmission periods. Under the given threshold of the control performance loss, a trade-off optimization problem is proposed. Finally, we give the solution to the optimization problem.
☆ DBPF: A Framework for Efficient and Robust Dynamic Bin-Picking
Efficiency and reliability are critical in robotic bin-picking as they directly impact the productivity of automated industrial processes. However, traditional approaches, demanding static objects and fixed collisions, lead to deployment limitations, operational inefficiencies, and process unreliability. This paper introduces a Dynamic Bin-Picking Framework (DBPF) that challenges traditional static assumptions. The DBPF endows the robot with the reactivity to pick multiple moving arbitrary objects while avoiding dynamic obstacles, such as the moving bin. Combined with scene-level pose generation, the proposed pose selection metric leverages the Tendency-Aware Manipulability Network optimizing suction pose determination. Heuristic task-specific designs like velocity-matching, dynamic obstacle avoidance, and the resight policy, enhance the picking success rate and reliability. Empirical experiments demonstrate the importance of these components. Our method achieves an average 84% success rate, surpassing the 60% of the most comparable baseline, crucially, with zero collisions. Further evaluations under diverse dynamic scenarios showcase DBPF's robust performance in dynamic bin-picking. Results suggest that our framework offers a promising solution for efficient and reliable robotic bin-picking under dynamics.
comment: 8 pages, 5 figures. This paper has been accepted by IEEE RA-L on 2024-03-24. See the supplementary video at youtube: https://youtu.be/n5af2VsKhkg
☆ Policy Gradient-based Model Free Optimal LQG Control with a Probabilistic Risk Constraint
In this paper, we investigate a model-free optimal control design that minimizes an infinite horizon average expected quadratic cost of states and control actions subject to a probabilistic risk or chance constraint using input-output data. In particular, we consider linear time-invariant systems and design an optimal controller within the class of linear state feedback control. Three different policy gradient (PG) based algorithms, natural policy gradient (NPG), Gauss-Newton policy gradient (GNPG), and deep deterministic policy gradient (DDPG), are developed, and compared with the optimal risk-neutral linear-quadratic regulator (LQR) and a scenario-based model predictive control (MPC) technique via numerical simulations. The convergence properties and the accuracy of all the algorithms are compared numerically. We also establish analytical convergence properties of the NPG and GNPG algorithms under the known model scenario, while the proof of convergence for the unknown model scenario is part of our ongoing work.
comment: Submitted to IEEE CDC2024
☆ A Blotto Game Approach to Ride-hailing Markets with Electric Vehicles
When a centrally operated ride-hailing company considers to enter a market already served by another company, it has to make a strategic decision about how to distribute its fleet among different regions in the area. This decision will be influenced by the market share the company can secure and the costs associated with charging the vehicles in each region, all while competing with the company already operating in the area. In this paper, we propose a Colonel Blotto-like game to model this decision-making. For the class of games that we study, we first prove the existence and uniqueness of a Nash Equilibrium. Subsequently, we provide its general characterization and present an algorithm for computing the ones in the feasible set's interior. Additionally, for a simplified scenario involving two regions, which would correspond to a city area with a downtown and a suburban region, we also provide a method to check for the equilibria on the feasible set's boundary. Finally, through a numerical case study, we illustrate the impact of charging prices on the position of the Nash equilibrium.
comment: Extended version of the paper accepted for presentation at the 2024 European Control Conference (ECC2024)
☆ Real-time Adaptation for Condition Monitoring Signal Prediction using Label-aware Neural Processes
Building a predictive model that rapidly adapts to real-time condition monitoring (CM) signals is critical for engineering systems/units. Unfortunately, many current methods suffer from a trade-off between representation power and agility in online settings. For instance, parametric methods that assume an underlying functional form for CM signals facilitate efficient online prediction updates. However, this simplification leads to vulnerability to model specifications and an inability to capture complex signals. On the other hand, approaches based on over-parameterized or non-parametric models can excel at explaining complex nonlinear signals, but real-time updates for such models pose a challenging task. In this paper, we propose a neural process-based approach that addresses this trade-off. It encodes available observations within a CM signal into a representation space and then reconstructs the signal's history and evolution for prediction. Once trained, the model can encode an arbitrary number of observations without requiring retraining, enabling on-the-spot real-time predictions along with quantified uncertainty and can be readily updated as more online data is gathered. Furthermore, our model is designed to incorporate qualitative information (i.e., labels) from individual units. This integration not only enhances individualized predictions for each unit but also enables joint inference for both signals and their associated labels. Numerical studies on both synthetic and real-world data in reliability engineering highlight the advantageous features of our model in real-time adaptation, enhanced signal prediction with uncertainty quantification, and joint prediction for labels and signals.
♻ ☆ Peak Estimation of Rational Systems using Convex Optimization
This paper presents algorithms that upper-bound the peak value of a state function along trajectories of a continuous-time system with rational dynamics. The finite-dimensional but nonconvex peak estimation problem is cast as a convex infinite-dimensional linear program in occupation measures. This infinite-dimensional program is then truncated into finite-dimensions using the moment-Sum-of-Squares (SOS) hierarchy of semidefinite programs. Prior work on treating rational dynamics using the moment-SOS approach involves clearing dynamics to common denominators or adding lifting variables to handle reciprocal terms under new equality constraints. Our solution method uses a sum-of-rational method based on absolute continuity of measures. The Moment-SOS truncations of our program possess lower computational complexity and (empirically demonstrated) higher accuracy of upper bounds on example systems as compared to prior approaches.
comment: 9 pages, 2 figures, 4 tables
♻ ☆ Robust Integral Consensus Control of Multi-Agent Networks Perturbed by Matched and Unmatched Disturbances: The Case of Directed Graphs
This work presents a new method to design consensus controllers for perturbed double integrator systems whose interconnection is described by a directed graph containing a rooted spanning tree. We propose new robust controllers to solve the consensus and synchronization problems when the systems are under the effects of matched and unmatched disturbances. In both problems, we present simple continuous controllers, whose integral actions allow us to handle the disturbances. A rigorous stability analysis based on Lyapunov's direct method for unperturbed networked systems is presented. To assess the performance of our result, a representative simulation study is presented.
♻ ☆ Time-Optimal Control for High-Order Chain-of-Integrators Systems with Full State Constraints and Arbitrary Terminal States (Part I, Extended Version)
Time-optimal control for high-order chain-of-integrators systems with full state constraints and arbitrary given terminal states remains a challenging problem in the optimal control theory domain, yet to be resolved. To enhance further comprehension of the problem, this paper establishes a novel notation system and theoretical framework, providing the switching manifold for high-order problems in the form of switching laws. Through deriving properties of switching laws on signs and dimension, this paper proposes a definite condition for time-optimal control. Guided by the developed theory, a trajectory planning method named the manifold-intercept method (MIM) is developed. The proposed MIM can plan time-optimal jerk-limited trajectories with full state constraints, and can also plan near-optimal non-chattering higher-order trajectories with negligible extra motion time compared to optimal profiles. Numerical results indicate that the proposed MIM outperforms all baselines in computational time, computational accuracy, and trajectory quality by a large gap.
♻ ☆ PowerSimulationsDynamics.jl -- An Open Source Modeling Package for Modern Power Systems with Inverter-Based Resources
In this paper we present the development of an open-source simulation toolbox, PowerSimulationsDynamics.jl, to study the dynamic response of power systems, focusing on the requirements to model systems with high penetrations of Inverter-Based Resources (IBRs). PowerSimulationsDynamics.jl is implemented in Julia and features a rich library of synchronous generator, inverter, and load models. In addition, it allows the study of quasi-static phasors and electromagnetic dq models that use a dynamic network representation. Case studies and validation exercises show that PowerSimulationsDynamics.jl results closely match other commercial and open-source simulation tools.
♻ ☆ Secure Control of Connected and Automated Vehicles Using Trust-Aware Robust Event-Triggered Control Barrier Functions
We address the security of a network of Connected and Automated Vehicles (CAVs) cooperating to safely navigate through a conflict area (e.g., traffic intersections, merging roadways, roundabouts). Previous studies have shown that such a network can be targeted by adversarial attacks causing traffic jams or safety violations ending in collisions. We focus on attacks targeting the V2X communication network used to share vehicle data and consider as well uncertainties due to noise in sensor measurements and communication channels. To combat these, motivated by recent work on the safe control of CAVs, we propose a trust-aware robust event-triggered decentralized control and coordination framework that can provably guarantee safety. We maintain a trust metric for each vehicle in the network computed based on their behavior and used to balance the tradeoff between conservativeness (when deeming every vehicle as untrustworthy) and guaranteed safety and security. It is important to highlight that our framework is invariant to the specific choice of the trust framework. Based on this framework, we propose an attack detection and mitigation scheme which has twofold benefits: (i) the trust framework is immune to false positives, and (ii) it provably guarantees safety against false positive cases. We use extensive simulations (in SUMO and CARLA) to validate the theoretical guarantees and demonstrate the efficacy of our proposed scheme to detect and mitigate adversarial attacks.
comment: arXiv admin note: substantial text overlap with arXiv:2305.16818
♻ ☆ Thermally-Resilient Soft Gripper for On-Orbit Operations IROS 2024
Research in soft manipulators has significantly enhanced object grasping capabilities, thanks to their adaptability to various shapes and sizes. Applying this technology to on-orbit servicing, especially during the capture and containment stages of active space debris removal missions, might offer a secure, adaptable, and cost-effective solution compared to the trend of increasing the degrees of freedom and complexity of the manipulator (e.g. ClearSpace, Astroscale). This work aims to conduct an experimental proof of concept, for which challenges such as radiation, vacuum, and microgravity are significant, but the predominant issue is ensuring effective operation in the extreme temperature swings, where flexible materials may exhibit cryogenic crystallization or drastic shifts in their elasticity. This work addresses this challenge through an initial stage of analytical modeling of the thermal dynamics inside the manipulator in orbit; which is then used for the development of a first experimental prototype tested with liquid nitrogen and heat guns. The multi-layered design for Low Earth Orbit (LEO) leverages the properties of TPU at low infill rates for lightweight inherent flexibility, silicone rubber ensuring structural integrity, PTFE (Teflon) for unparalleled thermal stability, and aerogel for insulation. The tendon-actuated servo-driven gripper is tested in the laboratory by varying the shape and size of objects during the grasping. The results, based on servomotor force metrics to assess the flexible manipulator's adaptability and object capture efficiency across temperature changes, affirm the concept's viability. Forces increase up to 220$\%$ in cryogenic conditions and decrease by no more than 50$\%$ at high temperatures.
comment: Submitted to IROS 2024
♻ ☆ Directionality-Aware Mixture Model Parallel Sampling for Efficient Linear Parameter Varying Dynamical System Learning
The Linear Parameter Varying Dynamical System (LPV-DS) is an effective approach that learns stable, time-invariant motion policies using statistical modeling and semi-definite optimization to encode complex motions for reactive robot control. Despite its strengths, the LPV-DS learning approach faces challenges in achieving a high model accuracy without compromising the computational efficiency. To address this, we introduce the Directionality-Aware Mixture Model (DAMM), a novel statistical model that applies the Riemannian metric on the n-sphere $\mathbb{S}^n$ to efficiently blend non-Euclidean directional data with $\mathbb{R}^m$ Euclidean states. Additionally, we develop a hybrid Markov chain Monte Carlo technique that combines Gibbs Sampling with Split/Merge Proposal, allowing for parallel computation to drastically speed up inference. Our extensive empirical tests demonstrate that LPV-DS integrated with DAMM achieves higher reproduction accuracy, better model efficiency, and near real-time/online learning compared to standard estimation methods on various datasets. Lastly, we demonstrate its suitability for incrementally learning multi-behavior policies in real-world robot experiments.
♻ ☆ Stable Reduced-Rank VAR Identification
The vector autoregression (VAR) has been widely used in system identification, econometrics, natural science, and many other areas. However, when the state dimension becomes large the parameter dimension explodes. So rank reduced modelling is attractive and is well developed. But a fundamental requirement in almost all applications is stability of the fitted model. And this has not been addressed in the rank reduced case. Here, we develop, for the first time, a closed-form formula for an estimator of a rank reduced transition matrix which is guaranteed to be stable. We show that our estimator is consistent and asymptotically statistically efficient and illustrate it in comparative simulations.
comment: 16 pages, 6 figures
♻ ☆ Stochastic Approximation with Delayed Updates: Finite-Time Rates under Markovian Sampling AISTATS
Motivated by applications in large-scale and multi-agent reinforcement learning, we study the non-asymptotic performance of stochastic approximation (SA) schemes with delayed updates under Markovian sampling. While the effect of delays has been extensively studied for optimization, the manner in which they interact with the underlying Markov process to shape the finite-time performance of SA remains poorly understood. In this context, our first main contribution is to show that under time-varying bounded delays, the delayed SA update rule guarantees exponentially fast convergence of the \emph{last iterate} to a ball around the SA operator's fixed point. Notably, our bound is \emph{tight} in its dependence on both the maximum delay $\tau_{max}$, and the mixing time $\tau_{mix}$. To achieve this tight bound, we develop a novel inductive proof technique that, unlike various existing delayed-optimization analyses, relies on establishing uniform boundedness of the iterates. As such, our proof may be of independent interest. Next, to mitigate the impact of the maximum delay on the convergence rate, we provide the first finite-time analysis of a delay-adaptive SA scheme under Markovian sampling. In particular, we show that the exponent of convergence of this scheme gets scaled down by $\tau_{avg}$, as opposed to $\tau_{max}$ for the vanilla delayed SA rule; here, $\tau_{avg}$ denotes the average delay across all iterations. Moreover, the adaptive scheme requires no prior knowledge of the delay sequence for step-size tuning. Our theoretical findings shed light on the finite-time effects of delays for a broad class of algorithms, including TD learning, Q-learning, and stochastic gradient descent under Markovian sampling.
comment: Accepted to the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024!
♻ ☆ Fault Localization for Buggy Deep Learning Framework Conversions in Image Recognition
When deploying Deep Neural Networks (DNNs), developers often convert models from one deep learning framework to another (e.g., TensorFlow to PyTorch). However, this process is error-prone and can impact target model accuracy. To identify the extent of such impact, we perform and briefly present a differential analysis against three DNNs widely used for image recognition (MobileNetV2, ResNet101, and InceptionV3) converted across four well-known deep learning frameworks (PyTorch, Keras, TensorFlow (TF), and TFLite), which revealed numerous model crashes and output label discrepancies of up to 100%. To mitigate such errors, we present a novel approach towards fault localization and repair of buggy deep learning framework conversions, focusing on pre-trained image recognition models. Our technique consists of four stages of analysis: 1) conversion tools, 2) model parameters, 3) model hyperparameters, and 4) graph representation. In addition, we propose various strategies towards fault repair of the faults detected. We implement our technique on top of the Apache TVM deep learning compiler, and we test it by conducting a preliminary fault localization analysis for the conversion of InceptionV3 from TF to TFLite. Our approach detected a fault in a common DNN converter tool, which introduced precision errors in weights, reducing model accuracy. After our fault localization, we repaired the issue, reducing our conversion error to zero.
comment: 5 pages, 3 figures, 1 table
♻ ☆ DeltaNN: Assessing the Impact of Computational Environment Parameters on the Performance of Image Recognition Models
Image recognition tasks typically use deep learning and require enormous processing power, thus relying on hardware accelerators like GPUs and TPUs for fast, timely processing. Failure in real-time image recognition tasks can occur due to sub-optimal mapping on hardware accelerators during model deployment, which may lead to timing uncertainty and erroneous behavior. Mapping on hardware accelerators is done using multiple software components like deep learning frameworks, compilers, and device libraries, that we refer to as the computational environment. Owing to the increased use of image recognition tasks in safety-critical applications like autonomous driving and medical imaging, it is imperative to assess their robustness to changes in the computational environment, as the impact of parameters like deep learning frameworks, compiler optimizations, and hardware devices on model performance and correctness is not yet well understood. In this paper we present a differential testing framework, DeltaNN, that allows us to assess the impact of different computational environment parameters on the performance of image recognition models during deployment, post training. DeltaNN generates different implementations of a given image recognition model for variations in environment parameters, namely, deep learning frameworks, compiler optimizations and hardware devices and analyzes differences in model performance as a result. Using DeltaNN, we conduct an empirical study of robustness analysis of three popular image recognition models using the ImageNet dataset. We report the impact in terms of misclassifications and inference time differences across different settings. In total, we observed up to 100% output label differences across deep learning frameworks, and up to 81% unexpected performance degradation in terms of inference time, when applying compiler optimizations.
comment: 11 pages, 10 figures, 2 tables
♻ ☆ Long Solution Times or Low Solution Quality: On Trade-Offs in Choosing a Power Flow Formulation for the Optimal Power Shutoff Problem
The Optimal Power Shutoff (OPS) problem is an optimization problem that makes power line de-energization decisions in order to reduce the risk of igniting a wildfire, while minimizing the load shed of customers. This problem, with DC linear power flow equations, has been used in many studies in recent years. However, using linear approximations for power flow when making decisions on the network topology is known to cause challenges with AC feasibility of the resulting network, as studied in the related contexts of optimal transmission switching or grid restoration planning. This paper explores the accuracy of the DC OPS formulation and the ability to recover an AC-feasible power flow solution after de-energization decisions are made. We also extend the OPS problem to include variants with the AC, Second-Order-Cone, and Network-Flow power flow equations, and compare them to the DC approximation with respect to solution quality and time. The results highlight that the DC approximation overestimates the amount of load that can be served, leading to poor de-energization decisions. The AC and SOC-based formulations are better, but prohibitively slow to solve for even modestly sized networks thus demonstrating the need for new solution methods with better trade-offs between computational time and solution quality.
♻ ☆ Investigating Robustness in Cyber-Physical Systems: Specification-Centric Analysis in the face of System Deviations
The adoption of cyber-physical systems (CPS) is on the rise in complex physical environments, encompassing domains such as autonomous vehicles, the Internet of Things (IoT), and smart cities. A critical attribute of CPS is robustness, denoting its capacity to operate safely despite potential disruptions and uncertainties in the operating environment. This paper proposes a novel specification-based robustness, which characterizes the effectiveness of a controller in meeting a specified system requirement, articulated through Signal Temporal Logic (STL) while accounting for possible deviations in the system. This paper also proposes the robustness falsification problem based on the definition, which involves identifying minor deviations capable of violating the specified requirement. We present an innovative two-layer simulation-based analysis framework designed to identify subtle robustness violations. To assess our methodology, we devise a series of benchmark problems wherein system parameters can be adjusted to emulate various forms of uncertainties and disturbances. Initial evaluations indicate that our falsification approach proficiently identifies robustness violations, providing valuable insights for comparing robustness between conventional and reinforcement learning (RL)-based controllers
comment: 12 pages
♻ ☆ On the Feedback Law in Stochastic Optimal Nonlinear Control
We consider the problem of nonlinear stochastic optimal control. This problem is thought to be fundamentally intractable owing to Bellman's ``curse of dimensionality". We present a result that shows that repeatedly solving an open-loop deterministic problem from the current state with progressively shorter horizons, similar to Model Predictive Control (MPC), results in a feedback policy that is $O(\epsilon^4)$ near to the true global stochastic optimal policy, \nxx{where $\epsilon$ is a perturbation parameter modulating the noise.} We show that the optimal deterministic feedback problem has a perturbation structure in that higher-order terms of the feedback law do not affect lower-order terms, and that this structure is lost in the optimal stochastic feedback problem. Consequently, solving the Stochastic Dynamic Programming problem is highly susceptible to noise, even when tractable, and in practice, the MPC-type feedback law offers superior performance even for stochastic systems.
comment: arXiv admin note: substantial text overlap with arXiv:2002.10505, arXiv:2002.09478
♻ ☆ Carbon Footprint Reduction for Sustainable Data Centers in Real-Time
As machine learning workloads significantly increase energy consumption, sustainable data centers with low carbon emissions are becoming a top priority for governments and corporations worldwide. This requires a paradigm shift in optimizing power consumption in cooling and IT loads, shifting flexible loads based on the availability of renewable energy in the power grid, and leveraging battery storage from the uninterrupted power supply in data centers, using collaborative agents. The complex association between these optimization strategies and their dependencies on variable external factors like weather and the power grid carbon intensity makes this a hard problem. Currently, a real-time controller to optimize all these goals simultaneously in a dynamic real-world setting is lacking. We propose a Data Center Carbon Footprint Reduction (DC-CFR) multi-agent Reinforcement Learning (MARL) framework that optimizes data centers for the multiple objectives of carbon footprint reduction, energy consumption, and energy cost. The results show that the DC-CFR MARL agents effectively resolved the complex interdependencies in optimizing cooling, load shifting, and energy storage in real-time for various locations under real-world dynamic weather and grid carbon intensity conditions. DC-CFR significantly outperformed the industry standard ASHRAE controller with a considerable reduction in carbon emissions (14.5%), energy usage (14.4%), and energy cost (13.7%) when evaluated over one year across multiple geographical regions.
♻ ☆ Identification of Energy Management Configuration Concepts from a Set of Pareto-optimal Solutions
Implementing resource efficient energy management systems in facilities and buildings becomes increasingly important in the transformation to a sustainable society. However, selecting a suitable configuration based on multiple, typically conflicting objectives, such as cost, robustness with respect to uncertainty of grid operation, or renewable energy utilization, is a difficult multi-criteria decision making problem. The recently developed concept identification technique can facilitate a decision maker by sorting configuration options into semantically meaningful groups (concepts). In this process, the partitioning of the objectives and design parameters into different sets (called description spaces) is a very important step. In this study we focus on utilizing the concept identification technique for finding relevant and viable energy management configurations from a very large data set of Pareto-optimal solutions. The data set consists of 20000 realistic Pareto-optimal building energy management configurations generated by a many-objective evolutionary optimization of a high quality Digital Twin energy management simulator. We analyze how the choice of description spaces, i.e., the partitioning of the objectives and parameters, impacts the type of information that can be extracted. We show that the decision maker can introduce constraints and biases into that process to meet expectations and preferences. The iterative approach presented in this work allows for the generation of valuable insights into trade-offs between specific objectives, and constitutes a powerful and flexible tool to support the decision making process when designing large and complex energy management systems.
comment: 18 pages, 8 figures, accepted at Energy Conversion and Management: X
♻ ☆ Hybrid physics-informed metabolic cybergenetics: process rates augmented with machine-learning surrogates informed by flux balance analysis
Metabolic cybergenetics is a promising concept that interfaces gene expression and cellular metabolism with computers for real-time dynamic metabolic control. The focus is on control at the transcriptional level, serving as a means to modulate intracellular metabolic fluxes. Recent strategies in this field have employed constraint-based dynamic models for process optimization, control, and estimation. However, this results in bilevel dynamic optimization problems, which pose considerable numerical and conceptual challenges. In this study, we present an alternative hybrid physics-informed dynamic modeling framework for metabolic cybergenetics, aimed at simplifying optimization, control, and estimation tasks. By utilizing machine-learning surrogates, our approach effectively embeds the physics of metabolic networks into the process rates of structurally simpler macro-kinetic models coupled with gene expression. These surrogates, informed by flux balance analysis, link the domains of manipulatable intracellular enzymes to metabolic exchange fluxes. This ensures that critical knowledge captured by the system's metabolic network is preserved. The resulting models can be integrated into metabolic cybergenetic schemes involving single-level optimizations. Additionally, the hybrid modeling approach maintains the number of system states at a necessary minimum, easing the burden of process monitoring and estimation. Our hybrid physics-informed metabolic cybergenetic framework is demonstrated using a computational case study on the optogenetically-assisted production of itaconate by $\textit{Escherichia coli}$.
comment: 25 pages, 10 figures, journal submission (reviewed/accepted version)
♻ ☆ Joint Ranging and Phase Offset Estimation for Multiple Drones using ADS-B Signatures
A new method for joint ranging and Phase Offset (PO) estimation of multiple drones/aircrafts is proposed in this paper. The proposed method employs the superimposed uncoordinated Automatic Dependent Surveillance Broadcast (ADS-B) packets broadcasted by drones/aircrafts for joint range and PO estimation. It jointly estimates range and PO prior to ADS-B packet decoding; thus, it can improve air safety when packet decoding is infeasible due to packet collision. Moreover, it enables coherent detection of ADS-B packets, which can result in more reliable multiple target tracking in aviation systems using cooperative sensors for detect and avoid (DAA). By minimizing the Kullback Leibler Divergence (KLD) statistical distance measure, we show that the received complex baseband signal coming from K uncoordinated drones corrupted by Additive White Gaussian Noise (AWGN) at a single antenna receiver can be approximated by an independent and identically distributed Gaussian Mixture (GM) with 2 power K mixture components in the two dimensional (2D) plane. While direct joint Maximum Likelihood Estimation (MLE) of range and PO from the derived GM Probability Density Function (PDF) leads to an intractable maximization, our proposed method employs the Expectation Maximization (EM) algorithm to estimate the modes of the 2D Gaussian mixture followed by a reordering estimation technique through combinatorial optimization to estimate range and PO. An extension to a multiple antenna receiver is also investigated in this paper. While the proposed estimator can estimate the range of multiple drones with a single receive antenna, a larger number of drones can be supported with higher accuracy by the use of multiple antennas at the receiver. The effectiveness of the proposed estimator is supported by simulation results. We show that the proposed estimator can jointly estimate the range of three drones accurately.
Robotics 25
☆ Single-Motor Robotic Gripper with Multi-Surface Fingers for Variable Grasping Configurations
This study proposes a novel robotic gripper with variable grasping configurations for grasping various objects. The fingers of the developed gripper incorporate multiple different surfaces. The gripper possesses the function of altering the finger surfaces facing a target object by rotating the fingers in its longitudinal direction. In the proposed design equipped with two fingers, the two fingers incorporate three and four surfaces, respectively, resulting in the nine available grasping configurations by the combination of these finger surfaces. The developed gripper is equipped with the functions of opening/closing its fingers for grasping and rotating its fingers to alter the grasping configuration -all achieved with a single motor. To enable the two motions using a single motor, this study introduces a self-motion switching mechanism utilizing magnets. This mechanism automatically transitions between gripper motions based on the direction of the motor rotation when the gripper is fully opened. In this state, rotating the motor towards closing initiates the finger closing action, while further opening the fingers from the fully opened state activates the finger rotation. This letter presents the gripper design, the mechanics of the self-motion switching mechanism, the control method, and the grasping configuration selection strategy. The performance of the gripper is experimentally demonstrated.
☆ Guessing human intentions to avoid dangerous situations in caregiving robots IROS
For robots to interact socially, they must interpret human intentions and anticipate their potential outcomes accurately. This is particularly important for social robots designed for human care, which may face potentially dangerous situations for people, such as unseen obstacles in their way, that should be avoided. This paper explores the Artificial Theory of Mind (ATM) approach to inferring and interpreting human intentions. We propose an algorithm that detects risky situations for humans, selecting a robot action that removes the danger in real time. We use the simulation-based approach to ATM and adopt the 'like-me' policy to assign intentions and actions to people. Using this strategy, the robot can detect and act with a high rate of success under time-constrained situations. The algorithm has been implemented as part of an existing robotics cognitive architecture and tested in simulation scenarios. Three experiments have been conducted to test the implementation's robustness, precision and real-time response, including a simulated scenario, a human-in-the-loop hybrid configuration and a real-world scenario.
comment: 8 pages, 6 figures. Submitted to IROS
☆ Combined Task and Motion Planning Via Sketch Decompositions (Extended Version with Supplementary Material)
The challenge in combined task and motion planning (TAMP) is the effective integration of a search over a combinatorial space, usually carried out by a task planner, and a search over a continuous configuration space, carried out by a motion planner. Using motion planners for testing the feasibility of task plans and filling out the details is not effective because it makes the geometrical constraints play a passive role. This work introduces a new interleaved approach for integrating the two dimensions of TAMP that makes use of sketches, a recent simple but powerful language for expressing the decomposition of problems into subproblems. A sketch has width 1 if it decomposes the problem into subproblems that can be solved greedily in linear time. In the paper, a general sketch is introduced for several classes of TAMP problems which has width 1 under suitable assumptions. While sketch decompositions have been developed for classical planning, they offer two important benefits in the context of TAMP. First, when a task plan is found to be unfeasible due to the geometric constraints, the combinatorial search resumes in a specific sub-problem. Second, the sampling of object configurations is not done once, globally, at the start of the search, but locally, at the start of each subproblem. Optimizations of this basic setting are also considered and experimental results over existing and new pick-and-place benchmarks are reported.
☆ M^3RS: Multi-robot, Multi-objective, and Multi-mode Routing and Scheduling
In this paper, we present a novel problem coined multi-robot, multi-objective, and multi-mode routing and scheduling (M^3RS). The formulation for M^3RS is introduced for time-bound multi-robot, multi-objective routing and scheduling missions where each task has multiple execution modes. Different execution modes have distinct resource consumption, associated execution time, and quality. M^3RS assigns the optimal sequence of tasks and the execution modes to each agent. The routes and associated modes depend on user preferences for different objective criteria. The need for M^3RS comes from multi-robot applications in which a trade-off between multiple criteria arises from different task execution modes. We use M^3RS for the application of multi-robot disinfection in public locations. The objectives considered for disinfection application are disinfection quality and number of tasks completed. A mixed-integer linear programming model is proposed for M^3RS. Then, a time-efficient column generation scheme is presented to tackle the issue of computation times for larger problem instances. The advantage of using multiple modes over fixed execution mode is demonstrated using experiments on synthetic data. The results suggest that M^3RS provides flexibility to the user in terms of available solutions and performs well in joint performance metrics. The application of the proposed problem is shown for a team of disinfection robots.} The videos for the experiments are available on the project website: https://sites.google.com/view/g-robot/m3rs/ .
☆ HT-LIP Model based Robust Control of Quadrupedal Robot Locomotion under Unknown Vertical Ground Motion
This paper presents a hierarchical control framework that enables robust quadrupedal locomotion on a dynamic rigid surface (DRS) with general and unknown vertical motions. The key novelty of the framework lies in its higher layer, which is a discrete-time, provably stabilizing footstep controller. The basis of the footstep controller is a new hybrid, time-varying, linear inverted pendulum (HT-LIP) model that is low-dimensional and accurately captures the essential robot dynamics during DRS locomotion. A new set of sufficient stability conditions are then derived to directly guide the controller design for ensuring the asymptotic stability of the HT-LIP model under general, unknown, vertical DRS motions. Further, the footstep controller is cast as a computationally efficient quadratic program that incorporates the proposed HT-LIP model and stability conditions. The middle layer takes the desired footstep locations generated by the higher layer as input to produce kinematically feasible full-body reference trajectories, which are then accurately tracked by a lower-layer torque controller. Hardware experiments on a Unitree Go1 quadrupedal robot confirm the robustness of the proposed framework under various unknown, aperiodic, vertical DRS motions and uncertainties (e.g., slippery and uneven surfaces, solid and liquid loads, and sudden pushes).
Legged Robot State Estimation within Non-inertial Environments
This paper investigates the robot state estimation problem within a non-inertial environment. The proposed state estimation approach relaxes the common assumption of static ground in the system modeling. The process and measurement models explicitly treat the movement of the non-inertial environments without requiring knowledge of its motion in the inertial frame or relying on GPS or sensing environmental landmarks. Further, the proposed state estimator is formulated as an invariant extended Kalman filter (InEKF) with the deterministic part of its process model obeying the group-affine property, leading to log-linear error dynamics. The observability analysis of the filter confirms that the robot's pose (i.e., position and orientation) and velocity relative to the non-inertial environment are observable. Hardware experiments on a humanoid robot moving on a rotating and translating treadmill demonstrate the high convergence rate and accuracy of the proposed InEKF even under significant treadmill pitch sway, as well as large estimation errors.
☆ KITchen: A Real-World Benchmark and Dataset for 6D Object Pose Estimation in Kitchen Environments
Despite the recent progress on 6D object pose estimation methods for robotic grasping, a substantial performance gap persists between the capabilities of these methods on existing datasets and their efficacy in real-world mobile manipulation tasks, particularly when robots rely solely on their monocular egocentric field of view (FOV). Existing real-world datasets primarily focus on table-top grasping scenarios, where a robotic arm is placed in a fixed position and the objects are centralized within the FOV of fixed external camera(s). Assessing performance on such datasets may not accurately reflect the challenges encountered in everyday mobile manipulation tasks within kitchen environments such as retrieving objects from higher shelves, sinks, dishwashers, ovens, refrigerators, or microwaves. To address this gap, we present Kitchen, a novel benchmark designed specifically for estimating the 6D poses of objects located in diverse positions within kitchen settings. For this purpose, we recorded a comprehensive dataset comprising around 205k real-world RGBD images for 111 kitchen objects captured in two distinct kitchens, utilizing one humanoid robot with its egocentric perspectives. Subsequently, we developed a semi-automated annotation pipeline, to streamline the labeling process of such datasets, resulting in the generation of 2D object labels, 2D object segmentation masks, and 6D object poses with minimized human effort. The benchmark, the dataset, and the annotation pipeline are available at https://kitchen-dataset.github.io/KITchen.
☆ Mixed-Initiative Human-Robot Teaming under Suboptimality with Online Bayesian Adaptation
For effective human-agent teaming, robots and other artificial intelligence (AI) agents must infer their human partner's abilities and behavioral response patterns and adapt accordingly. Most prior works make the unrealistic assumption that one or more teammates can act near-optimally. In real-world collaboration, humans and autonomous agents can be suboptimal, especially when each only has partial domain knowledge. In this work, we develop computational modeling and optimization techniques for enhancing the performance of suboptimal human-agent teams, where the human and the agent have asymmetric capabilities and act suboptimally due to incomplete environmental knowledge. We adopt an online Bayesian approach that enables a robot to infer people's willingness to comply with its assistance in a sequential decision-making game. Our user studies show that user preferences and team performance indeed vary with robot intervention styles, and our approach for mixed-initiative collaborations enhances objective team performance ($p<.001$) and subjective measures, such as user's trust ($p<.001$) and perceived likeability of the robot ($p<.001$).
comment: 8 pages, 4 pages for supplementary
☆ Realtime Robust Shape Estimation of Deformable Linear Object ICRA 2024
Realtime shape estimation of continuum objects and manipulators is essential for developing accurate planning and control paradigms. The existing methods that create dense point clouds from camera images, and/or use distinguishable markers on a deformable body have limitations in realtime tracking of large continuum objects/manipulators. The physical occlusion of markers can often compromise accurate shape estimation. We propose a robust method to estimate the shape of linear deformable objects in realtime using scattered and unordered key points. By utilizing a robust probability-based labeling algorithm, our approach identifies the true order of the detected key points and then reconstructs the shape using piecewise spline interpolation. The approach only relies on knowing the number of the key points and the interval between two neighboring points. We demonstrate the robustness of the method when key points are partially occluded. The proposed method is also integrated into a simulation in Unity for tracking the shape of a cable with a length of 1m and a radius of 5mm. The simulation results show that our proposed approach achieves an average length error of 1.07% over the continuum's centerline and an average cross-section error of 2.11mm. The real-world experiments of tracking and estimating a heavy-load cable prove that the proposed approach is robust under occlusion and complex entanglement scenarios.
comment: This paper has been accepted to IEEE ICRA 2024 as a contributed paper
☆ CG-SLAM: Efficient Dense RGB-D SLAM in a Consistent Uncertainty-aware 3D Gaussian Field
Recently neural radiance fields (NeRF) have been widely exploited as 3D representations for dense simultaneous localization and mapping (SLAM). Despite their notable successes in surface modeling and novel view synthesis, existing NeRF-based methods are hindered by their computationally intensive and time-consuming volume rendering pipeline. This paper presents an efficient dense RGB-D SLAM system, i.e., CG-SLAM, based on a novel uncertainty-aware 3D Gaussian field with high consistency and geometric stability. Through an in-depth analysis of Gaussian Splatting, we propose several techniques to construct a consistent and stable 3D Gaussian field suitable for tracking and mapping. Additionally, a novel depth uncertainty model is proposed to ensure the selection of valuable Gaussian primitives during optimization, thereby improving tracking efficiency and accuracy. Experiments on various datasets demonstrate that CG-SLAM achieves superior tracking and mapping performance with a notable tracking speed of up to 15 Hz. We will make our source code publicly available. Project page: https://zju3dv.github.io/cg-slam.
comment: Project Page: https://zju3dv.github.io/cg-slam
☆ Are NeRFs ready for autonomous driving? Towards closing the real-to-simulation gap
Neural Radiance Fields (NeRFs) have emerged as promising tools for advancing autonomous driving (AD) research, offering scalable closed-loop simulation and data augmentation capabilities. However, to trust the results achieved in simulation, one needs to ensure that AD systems perceive real and rendered data in the same way. Although the performance of rendering methods is increasing, many scenarios will remain inherently challenging to reconstruct faithfully. To this end, we propose a novel perspective for addressing the real-to-simulated data gap. Rather than solely focusing on improving rendering fidelity, we explore simple yet effective methods to enhance perception model robustness to NeRF artifacts without compromising performance on real data. Moreover, we conduct the first large-scale investigation into the real-to-simulated data gap in an AD setting using a state-of-the-art neural rendering technique. Specifically, we evaluate object detectors and an online mapping model on real and simulated data, and study the effects of different pre-training strategies. Our results show notable improvements in model robustness to simulated data, even improving real-world performance in some cases. Last, we delve into the correlation between the real-to-simulated gap and image reconstruction metrics, identifying FID and LPIPS as strong indicators.
☆ RPMArt: Towards Robust Perception and Manipulation for Articulated Objects IROS 2024
Articulated objects are commonly found in daily life. It is essential that robots can exhibit robust perception and manipulation skills for articulated objects in real-world robotic applications. However, existing methods for articulated objects insufficiently address noise in point clouds and struggle to bridge the gap between simulation and reality, thus limiting the practical deployment in real-world scenarios. To tackle these challenges, we propose a framework towards Robust Perception and Manipulation for Articulated Objects (RPMArt), which learns to estimate the articulation parameters and manipulate the articulation part from the noisy point cloud. Our primary contribution is a Robust Articulation Network (RoArtNet) that is able to predict both joint parameters and affordable points robustly by local feature learning and point tuple voting. Moreover, we introduce an articulation-aware classification scheme to enhance its ability for sim-to-real transfer. Finally, with the estimated affordable point and articulation joint constraint, the robot can generate robust actions to manipulate articulated objects. After learning only from synthetic data, RPMArt is able to transfer zero-shot to real-world articulated objects. Experimental results confirm our approach's effectiveness, with our framework achieving state-of-the-art performance in both noise-added simulation and real-world environments. The code and data will be open-sourced for reproduction. More results are published on the project website at https://r-pmart.github.io .
comment: 8 pages, 7 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024), project website at https://r-pmart.github.io
☆ MQE: Unleashing the Power of Interaction with Multi-agent Quadruped Environment
The advent of deep reinforcement learning (DRL) has significantly advanced the field of robotics, particularly in the control and coordination of quadruped robots. However, the complexity of real-world tasks often necessitates the deployment of multi-robot systems capable of sophisticated interaction and collaboration. To address this need, we introduce the Multi-agent Quadruped Environment (MQE), a novel platform designed to facilitate the development and evaluation of multi-agent reinforcement learning (MARL) algorithms in realistic and dynamic scenarios. MQE emphasizes complex interactions between robots and objects, hierarchical policy structures, and challenging evaluation scenarios that reflect real-world applications. We present a series of collaborative and competitive tasks within MQE, ranging from simple coordination to complex adversarial interactions, and benchmark state-of-the-art MARL algorithms. Our findings indicate that hierarchical reinforcement learning can simplify task learning, but also highlight the need for advanced algorithms capable of handling the intricate dynamics of multi-agent interactions. MQE serves as a stepping stone towards bridging the gap between simulation and practical deployment, offering a rich environment for future research in multi-agent systems and robot learning. For open-sourced code and more details of MQE, please refer to https://ziyanx02.github.io/multiagent-quadruped-environment/ .
comment: Open-source code is available at https://github.com/ziyanx02/multiagent-quadruped-environment
☆ Robust-Locomotion-by-Logic: Perturbation-Resilient Bipedal Locomotion via Signal Temporal Logic Guided Model Predictive Control
This study introduces a robust planning framework that utilizes a model predictive control (MPC) approach, enhanced by incorporating signal temporal logic (STL) specifications. This marks the first-ever study to apply STL-guided trajectory optimization for bipedal locomotion, specifically designed to handle both translational and orientational perturbations. Existing recovery strategies often struggle with reasoning complex task logic and evaluating locomotion robustness systematically, making them susceptible to failures caused by inappropriate recovery strategies or lack of robustness. To address these issues, we design an analytical robustness metric for bipedal locomotion and quantify this metric using STL specifications, which guide the generation of recovery trajectories to achieve maximum locomotion robustness. To enable safe and computational-efficient crossed-leg maneuver, we design data-driven self-leg-collision constraints that are $1000$ times faster than the traditional inverse-kinematics-based approach. Our framework outperforms a state-of-the-art locomotion controller, a standard MPC without STL, and a linear-temporal-logic-based planner in a high-fidelity dynamic simulation, especially in scenarios involving crossed-leg maneuvers. Additionally, the Cassie bipedal robot achieves robust performance under horizontal and orientational perturbations such as those observed in ship motions. These environments are validated in simulations and deployed on hardware. Furthermore, our proposed method demonstrates versatility on stepping stones and terrain-agnostic features on inclined terrains.
♻ ☆ Make a Donut: Hierarchical EMD-Space Planning for Zero-Shot Deformable Manipulation with Tools
Deformable object manipulation stands as one of the most captivating yet formidable challenges in robotics. While previous techniques have predominantly relied on learning latent dynamics through demonstrations, typically represented as either particles or images, there exists a pertinent limitation: acquiring suitable demonstrations, especially for long-horizon tasks, can be elusive. Moreover, basing learning entirely on demonstrations can hamper the model's ability to generalize beyond the demonstrated tasks. In this work, we introduce a demonstration-free hierarchical planning approach capable of tackling intricate long-horizon tasks without necessitating any training. We employ large language models (LLMs) to articulate a high-level, stage-by-stage plan corresponding to a specified task. For every individual stage, the LLM provides both the tool's name and the Python code to craft intermediate subgoal point clouds. With the tool and subgoal for a particular stage at our disposal, we present a granular closed-loop model predictive control strategy. This leverages Differentiable Physics with Point-to-Point correspondence (DiffPhysics-P2P) loss in the earth mover distance (EMD) space, applied iteratively. Experimental findings affirm that our technique surpasses multiple benchmarks in dough manipulation, spanning both short and long horizons. Remarkably, our model demonstrates robust generalization capabilities to novel and previously unencountered complex tasks without any preliminary demonstrations. We further substantiate our approach with experimental trials on real-world robotic platforms. Our project page: https://qq456cvb.github.io/projects/donut.
comment: 8 pages
♻ ☆ Development and Evaluation of a Learning-based Model for Real-time Haptic Texture Rendering
Current Virtual Reality (VR) environments lack the rich haptic signals that humans experience during real-life interactions, such as the sensation of texture during lateral movement on a surface. Adding realistic haptic textures to VR environments requires a model that generalizes to variations of a user's interaction and to the wide variety of existing textures in the world. Current methodologies for haptic texture rendering exist, but they usually develop one model per texture, resulting in low scalability. We present a deep learning-based action-conditional model for haptic texture rendering and evaluate its perceptual performance in rendering realistic texture vibrations through a multi part human user study. This model is unified over all materials and uses data from a vision-based tactile sensor (GelSight) to render the appropriate surface conditioned on the user's action in real time. For rendering texture, we use a high-bandwidth vibrotactile transducer attached to a 3D Systems Touch device. The result of our user study shows that our learning-based method creates high-frequency texture renderings with comparable or better quality than state-of-the-art methods without the need for learning a separate model per texture. Furthermore, we show that the method is capable of rendering previously unseen textures using a single GelSight image of their surface.
comment: Accepted for publication in IEEE Transactions on Haptics 2024. 12 pages, 8 figures
♻ ☆ Generative Graphical Inverse Kinematics
Quickly and reliably finding accurate inverse kinematics (IK) solutions remains a challenging problem for many robot manipulators. Existing numerical solvers are broadly applicable but typically only produce a single solution and rely on local search techniques to minimize nonconvex objective functions. More recent learning-based approaches that approximate the entire feasible set of solutions have shown promise as a means to generate multiple fast and accurate IK results in parallel. However, existing learning-based techniques have a significant drawback: each robot of interest requires a specialized model that must be trained from scratch. To address this key shortcoming, we propose a novel distance-geometric robot representation coupled with a graph structure that allows us to leverage the sample efficiency of Euclidean equivariant functions and the generalizability of graph neural networks (GNNs). Our approach is generative graphical inverse kinematics (GGIK), the first learned IK solver able to accurately and efficiently produce a large number of diverse solutions in parallel while also displaying the ability to generalize -- a single learned model can be used to produce IK solutions for a variety of different robots. When compared to several other learned IK methods, GGIK provides more accurate solutions with the same amount of data. GGIK can generalize reasonably well to robot manipulators unseen during training. Additionally, GGIK can learn a constrained distribution that encodes joint limits and scales efficiently to larger robots and a high number of sampled solutions. Finally, GGIK can be used to complement local IK solvers by providing reliable initializations for a local optimization process.
comment: Submitted to IEEE Transactions on Robotics, June 2023
♻ ☆ SG-Bot: Object Rearrangement via Coarse-to-Fine Robotic Imagination on Scene Graphs ICRA 2024
Object rearrangement is pivotal in robotic-environment interactions, representing a significant capability in embodied AI. In this paper, we present SG-Bot, a novel rearrangement framework that utilizes a coarse-to-fine scheme with a scene graph as the scene representation. Unlike previous methods that rely on either known goal priors or zero-shot large models, SG-Bot exemplifies lightweight, real-time, and user-controllable characteristics, seamlessly blending the consideration of commonsense knowledge with automatic generation capabilities. SG-Bot employs a three-fold procedure--observation, imagination, and execution--to adeptly address the task. Initially, objects are discerned and extracted from a cluttered scene during the observation. These objects are first coarsely organized and depicted within a scene graph, guided by either commonsense or user-defined criteria. Then, this scene graph subsequently informs a generative model, which forms a fine-grained goal scene considering the shape information from the initial scene and object semantics. Finally, for execution, the initial and envisioned goal scenes are matched to formulate robotic action policies. Experimental results demonstrate that SG-Bot outperforms competitors by a large margin.
comment: ICRA 2024 accepted. Project website: https://sites.google.com/view/sg-bot
♻ ☆ EnduRL: Enhancing Safety, Stability, and Efficiency of Mixed Traffic Under Real-World Perturbations Via Reinforcement Learning
Human-driven vehicles (HVs) amplify naturally occurring perturbations in traffic, leading to congestion--a major contributor to increased fuel consumption, higher collision risks, and reduced road capacity utilization. While previous research demonstrates that Robot Vehicles (RVs) can be leveraged to mitigate these issues, most such studies rely on simulations with simplistic models of human car-following behaviors. In this work, we analyze real-world driving trajectories and extract a wide range of acceleration profiles. We then incorporates these profiles into simulations for training RVs to mitigate congestion. We evaluate the safety, efficiency, and stability of mixed traffic via comprehensive experiments conducted in two mixed traffic environments (Ring and Bottleneck) at various traffic densities, configurations, and RV penetration rates. The results show that under real-world perturbations, prior RV controllers experience performance degradation on all three objectives (sometimes even lower than 100% HVs). To address this, we introduce a reinforcement learning based RV that employs a congestion stage classifier to optimize the safety, efficiency, and stability of mixed traffic. Our RVs demonstrate significant improvements: safety by up to 66%, efficiency by up to 54%, and stability by up to 97%.
♻ ☆ Soft finger rotational stability for precision grasps IROS24
Soft robotic fingers can safely grasp fragile or variable form objects, but their force capacity is limited, especially with less contact area: precision grasps and when objects are smaller or not spherical. Current research is improving force capacity through mechanical design by increasing contact area or stiffness, typically without models which explain soft finger force limitations. To address this, this paper considers two types of soft grip failure, slip and dynamic rotational stability. For slip, the validity of a Coulomb model investigated, identifying the effect of contact area, pressure, and relative pose. For rotational stability, bulk linear stiffness of the fingers is used to develop conditions for dynamic stability and identify when rotation leads to slip. Together, these models suggest contact area improves force capacity by increasing transverse stiffness and normal force. The models are validated on pneumatic fingers, both custom PneuNets-based and commercially available. The models are used to find grip parameters which increase force capacity without failure.
comment: Submitted IROS24
♻ ☆ Combining Sampling- and Gradient-based Planning for Contact-rich Manipulation ICRA24
Planning over discontinuous dynamics is needed for robotics tasks like contact-rich manipulation, which presents challenges in the numerical stability and speed of planning methods when either neural network or analytical models are used. On the one hand, sampling-based planners require higher sample complexity in high-dimensional problems and cannot describe safety constraints such as force limits. On the other hand, gradient-based solvers can suffer from local optima and convergence issues when the Hessian is poorly conditioned. We propose a planning method with both sampling- and gradient-based elements, using the Cross-entropy Method to initialize a gradient-based solver, providing better search over local minima and the ability to handle explicit constraints. We show the approach allows smooth, stable contact-rich planning for an impedance-controlled robot making contact with a stiff environment, benchmarking against gradient-only MPC and CEM.
comment: Submitted ICRA24. Video available at https://youtu.be/COqR90392Kw Code available at https://gitlab.cc-asp.fraunhofer.de/hanikevi/contact_mpc
♻ ☆ SwarmPRM: Probabilistic Roadmap Motion Planning for Large-Scale Swarm Robotic Systems IROS 2024
Large-scale swarm robotic systems consisting of numerous cooperative agents show considerable promise for performing autonomous tasks across various sectors. Nonetheless, traditional motion planning approaches often face a trade-off between scalability and solution quality due to the exponential growth of the joint state space of robots. In response, this work proposes SwarmPRM, a hierarchical, scalable, computationally efficient, and risk-aware sampling-based motion planning approach for large-scale swarm robots. SwarmPRM utilizes a Gaussian Mixture Model (GMM) to represent the swarm's macroscopic state and constructs a Probabilistic Roadmap in Gaussian space, referred to as the Gaussian roadmap, to generate a transport trajectory of GMM. This trajectory is then followed by each robot at the microscopic stage. To enhance trajectory safety, SwarmPRM incorporates the conditional value-at-risk (CVaR) in the collision checking process to impart the property of risk awareness to the constructed Gaussian roadmap. SwarmPRM then crafts a linear programming formulation to compute the optimal GMM transport trajectory within this roadmap. Extensive simulations demonstrate that SwarmPRM outperforms state-of-the-art methods in computational efficiency, scalability, and trajectory quality while offering the capability to adjust the risk tolerance of generated trajectories.
comment: Submitted to IROS 2024
♻ ☆ DRL-Based Trajectory Tracking for Motion-Related Modules in Autonomous Driving
Autonomous driving systems are always built on motion-related modules such as the planner and the controller. An accurate and robust trajectory tracking method is indispensable for these motion-related modules as a primitive routine. Current methods often make strong assumptions about the model such as the context and the dynamics, which are not robust enough to deal with the changing scenarios in a real-world system. In this paper, we propose a Deep Reinforcement Learning (DRL)-based trajectory tracking method for the motion-related modules in autonomous driving systems. The representation learning ability of DL and the exploration nature of RL bring strong robustness and improve accuracy. Meanwhile, it enhances versatility by running the trajectory tracking in a model-free and data-driven manner. Through extensive experiments, we demonstrate both the efficiency and effectiveness of our method compared to current methods. Code and documentation are released to facilitate both further research and industrial deployment.
comment: Technical report. Code: https://github.com/MARMOTatZJU/drl-based-trajectory-tracking Documentation: https://drl-based-trajectory-tracking.readthedocs.io
♻ ☆ A Number Sense as an Emergent Property of the Manipulating Brain
The ability to understand and manipulate numbers and quantities emerges during childhood, but the mechanism through which humans acquire and develop this ability is still poorly understood. We explore this question through a model, assuming that the learner is able to pick up and place small objects from, and to, locations of its choosing, and will spontaneously engage in such undirected manipulation. We further assume that the learner's visual system will monitor the changing arrangements of objects in the scene and will learn to predict the effects of each action by comparing perception with a supervisory signal from the motor system. We model perception using standard deep networks for feature extraction and classification, and gradient descent learning. Our main finding is that, from learning the task of action prediction, an unexpected image representation emerges exhibiting regularities that foreshadow the perception and representation of numbers and quantity. These include distinct categories for zero and the first few natural numbers, a strict ordering of the numbers, and a one-dimensional signal that correlates with numerical quantity. As a result, our model acquires the ability to estimate numerosity, i.e. the number of objects in the scene, as well as subitization, i.e. the ability to recognize at a glance the exact number of objects in small scenes. Remarkably, subitization and numerosity estimation extrapolate to scenes containing many objects, far beyond the three objects used during training. We conclude that important aspects of a facility with numbers and quantities may be learned with supervision from a simple pre-training task. Our observations suggest that cross-modal learning is a powerful learning mechanism that may be harnessed in artificial intelligence.
comment: 16 pages, 5 figures, 15 supplemental figures
♻ ☆ Effective Integration of Weighted Cost-to-go and Conflict Heuristic within Suboptimal CBS AAAI 2023
Conflict-Based Search (CBS) is a popular multi-agent path finding (MAPF) solver that employs a low-level single agent planner and a high-level constraint tree to resolve conflicts. The vast majority of modern MAPF solvers focus on improving CBS by reducing the size of this tree through various strategies with few methods modifying the low level planner. Typically low level planners in existing CBS methods use an unweighted cost-to-go heuristic, with suboptimal CBS methods also using a conflict heuristic to help the high level search. In this paper, we show that, contrary to prevailing CBS beliefs, a weighted cost-to-go heuristic can be used effectively alongside the conflict heuristic in two possible variants. In particular, one of these variants can obtain large speedups, 2-100x, across several scenarios and suboptimal CBS methods. Importantly, we discover that performance is related not to the weighted cost-to-go heuristic but rather to the relative conflict heuristic weight's ability to effectively balance low-level and high-level work. Additionally, to the best of our knowledge, we show the first theoretical relation of prioritized planning and bounded suboptimal CBS and demonstrate that our methods are their natural generalization. Update March 2024: We found that the relative speedup decreases to around 1.2-10x depending on how the conflict heuristic is computed (see appendix for more details).
comment: Published in AAAI 2023
Artificial Intelligence 69
☆ Artificial Neural Microcircuits as Building Blocks: Concept and Challenges
Artificial Neural Networks (ANNs) are one of the most widely employed forms of bio-inspired computation. However the current trend is for ANNs to be structurally homogeneous. Furthermore, this structural homogeneity requires the application of complex training and learning tools that produce application specific ANNs, susceptible to pitfalls such as overfitting. In this paper, an new approach is explored, inspired by the role played in biology by Neural Microcircuits, the so called ``fundamental processing elements'' of organic nervous systems. How large neural networks, particularly Spiking Neural Networks (SNNs) can be assembled using Artificial Neural Microcircuits (ANMs), intended as off-the-shelf components, is articulated; the results of initial work to produce a catalogue of such Microcircuits though the use of Novelty Search is shown; followed by efforts to expand upon this initial work, including a discussion of challenges uncovered during these efforts and explorations of methods by which they might be overcome.
comment: 12 pages, 31 figures, 3 tables, submitted to A-Life Journal for review
☆ Large Language Models in Biomedical and Health Informatics: A Bibliometric Review
Large Language Models (LLMs) have rapidly become important tools in Biomedical and Health Informatics (BHI), enabling new ways to analyze data, treat patients, and conduct research. This bibliometric review aims to provide a panoramic view of how LLMs have been used in BHI by examining research articles and collaboration networks from 2022 to 2023. It further explores how LLMs can improve Natural Language Processing (NLP) applications in various BHI areas like medical diagnosis, patient engagement, electronic health record management, and personalized medicine. To do this, our bibliometric review identifies key trends, maps out research networks, and highlights major developments in this fast-moving field. Lastly, it discusses the ethical concerns and practical challenges of using LLMs in BHI, such as data privacy and reliable medical recommendations. Looking ahead, we consider how LLMs could further transform biomedical research as well as healthcare delivery and patient outcomes. This comprehensive review serves as a resource for stakeholders in healthcare, including researchers, clinicians, and policymakers, to understand the current state and future potential of LLMs in BHI.
comment: 50 pages, 7 figures, 4 tables
☆ Guessing human intentions to avoid dangerous situations in caregiving robots IROS
For robots to interact socially, they must interpret human intentions and anticipate their potential outcomes accurately. This is particularly important for social robots designed for human care, which may face potentially dangerous situations for people, such as unseen obstacles in their way, that should be avoided. This paper explores the Artificial Theory of Mind (ATM) approach to inferring and interpreting human intentions. We propose an algorithm that detects risky situations for humans, selecting a robot action that removes the danger in real time. We use the simulation-based approach to ATM and adopt the 'like-me' policy to assign intentions and actions to people. Using this strategy, the robot can detect and act with a high rate of success under time-constrained situations. The algorithm has been implemented as part of an existing robotics cognitive architecture and tested in simulation scenarios. Three experiments have been conducted to test the implementation's robustness, precision and real-time response, including a simulated scenario, a human-in-the-loop hybrid configuration and a real-world scenario.
comment: 8 pages, 6 figures. Submitted to IROS
☆ Engineering Safety Requirements for Autonomous Driving with Large Language Models
Changes and updates in the requirement artifacts, which can be frequent in the automotive domain, are a challenge for SafetyOps. Large Language Models (LLMs), with their impressive natural language understanding and generating capabilities, can play a key role in automatically refining and decomposing requirements after each update. In this study, we propose a prototype of a pipeline of prompts and LLMs that receives an item definition and outputs solutions in the form of safety requirements. This pipeline also performs a review of the requirement dataset and identifies redundant or contradictory requirements. We first identified the necessary characteristics for performing HARA and then defined tests to assess an LLM's capability in meeting these criteria. We used design science with multiple iterations and let experts from different companies evaluate each cycle quantitatively and qualitatively. Finally, the prototype was implemented at a case company and the responsible team evaluated its efficiency.
comment: Accepted in 32nd IEEE International Requirements Engineering 2024 conference, Iceland
☆ AVicuna: Audio-Visual LLM with Interleaver and Context-Boundary Alignment for Temporal Referential Dialogue
In everyday communication, humans frequently use speech and gestures to refer to specific areas or objects, a process known as Referential Dialogue (RD). While prior studies have investigated RD through Large Language Models (LLMs) or Large Multimodal Models (LMMs) in static contexts, the exploration of Temporal Referential Dialogue (TRD) within audio-visual media remains limited. Two primary challenges hinder progress in this field: (1) the absence of comprehensive, untrimmed audio-visual video datasets with precise temporal annotations, and (2) the need for methods to integrate complex temporal auditory and visual cues effectively. To address these challenges, we introduce a novel framework to generate PU-VALOR, an extensive audio-visual dataset comprising over 114,000 untrimmed videos with accurate temporal demarcations. We also present AVicuna, featuring an Audio-Visual Tokens Interleaver (AVTI) that ensures the temporal alignment of audio-visual information. Additionally, we develop the A5-222K dataset, encompassing more than 200,000 audio-text pairings, to facilitate the audio and text alignments. Our experiments demonstrate that AVicuna can effectively handle TRD in audio-visual videos and achieve state-of-the-art performance on various audio-visual video understanding tasks, particularly in untrimmed videos. We further investigate the optimal audio-interleaving rate for interleaved audio-visual inputs, which maximizes performance on the Audio-Visual Event Dense Localization task.
☆ L-MAE: Longitudinal masked auto-encoder with time and severity-aware encoding for diabetic retinopathy progression prediction
Pre-training strategies based on self-supervised learning (SSL) have proven to be effective pretext tasks for many downstream tasks in computer vision. Due to the significant disparity between medical and natural images, the application of typical SSL is not straightforward in medical imaging. Additionally, those pretext tasks often lack context, which is critical for computer-aided clinical decision support. In this paper, we developed a longitudinal masked auto-encoder (MAE) based on the well-known Transformer-based MAE. In particular, we explored the importance of time-aware position embedding as well as disease progression-aware masking. Taking into account the time between examinations instead of just scheduling them offers the benefit of capturing temporal changes and trends. The masking strategy, for its part, evolves during follow-up to better capture pathological changes, ensuring a more accurate assessment of disease progression. Using OPHDIAT, a large follow-up screening dataset targeting diabetic retinopathy (DR), we evaluated the pre-trained weights on a longitudinal task, which is to predict the severity label of the next visit within 3 years based on the past time series examinations. Our results demonstrated the relevancy of both time-aware position embedding and masking strategies based on disease progression knowledge. Compared to popular baseline models and standard longitudinal Transformers, these simple yet effective extensions significantly enhance the predictive ability of deep classification models.
☆ Out-of-Distribution Detection via Deep Multi-Comprehension Ensemble
Recent research underscores the pivotal role of the Out-of-Distribution (OOD) feature representation field scale in determining the efficacy of models in OOD detection. Consequently, the adoption of model ensembles has emerged as a prominent strategy to augment this feature representation field, capitalizing on anticipated model diversity. However, our introduction of novel qualitative and quantitative model ensemble evaluation methods, specifically Loss Basin/Barrier Visualization and the Self-Coupling Index, reveals a critical drawback in existing ensemble methods. We find that these methods incorporate weights that are affine-transformable, exhibiting limited variability and thus failing to achieve the desired diversity in feature representation. To address this limitation, we elevate the dimensions of traditional model ensembles, incorporating various factors such as different weight initializations, data holdout, etc., into distinct supervision tasks. This innovative approach, termed Multi-Comprehension (MC) Ensemble, leverages diverse training tasks to generate distinct comprehensions of the data and labels, thereby extending the feature representation field. Our experimental results demonstrate the superior performance of the MC Ensemble strategy in OOD detection compared to both the naive Deep Ensemble method and a standalone model of comparable size. This underscores the effectiveness of our proposed approach in enhancing the model's capability to detect instances outside its training distribution.
☆ On machine learning analysis of atomic force microscopy images for image classification, sample surface recognition
Atomic force microscopy (AFM or SPM) imaging is one of the best matches with machine learning (ML) analysis among microscopy techniques. The digital format of AFM images allows for direct utilization in ML algorithms without the need for additional processing. Additionally, AFM enables the simultaneous imaging of distributions of over a dozen different physicochemical properties of sample surfaces, a process known as multidimensional imaging. While this wealth of information can be challenging to analyze using traditional methods, ML provides a seamless approach to this task. However, the relatively slow speed of AFM imaging poses a challenge in applying deep learning methods broadly used in image recognition. This Prospective is focused on ML recognition/classification when using a relatively small number of AFM images, small database. We discuss ML methods other than popular deep-learning neural networks. The described approach has already been successfully used to analyze and classify the surfaces of biological cells. It can be applied to recognize medical images, specific material processing, in forensic studies, even to identify the authenticity of arts. A general template for ML analysis specific to AFM is suggested, with a specific example of the identification of cell phenotype. Special attention is given to the analysis of the statistical significance of the obtained results, an important feature that is often overlooked in papers dealing with machine learning. A simple method for finding statistical significance is also described.
comment: perspective; mini-review; method description; Physical Chemistry Chemical Physics (PCCP) in press, 2024
☆ Cyber-Security Knowledge Graph Generation by Hierarchical Nonnegative Matrix Factorization
Much of human knowledge in cybersecurity is encapsulated within the ever-growing volume of scientific papers. As this textual data continues to expand, the importance of document organization methods becomes increasingly crucial for extracting actionable insights hidden within large text datasets. Knowledge Graphs (KGs) serve as a means to store factual information in a structured manner, providing explicit, interpretable knowledge that includes domain-specific information from the cybersecurity scientific literature. One of the challenges in constructing a KG from scientific literature is the extraction of ontology from unstructured text. In this paper, we address this topic and introduce a method for building a multi-modal KG by extracting structured ontology from scientific papers. We demonstrate this concept in the cybersecurity domain. One modality of the KG represents observable information from the papers, such as the categories in which they were published or the authors. The second modality uncovers latent (hidden) patterns of text extracted through hierarchical and semantic non-negative matrix factorization (NMF), such as named entities, topics or clusters, and keywords. We illustrate this concept by consolidating more than two million scientific papers uploaded to arXiv into the cyber-domain, using hierarchical and semantic NMF, and by building a cyber-domain-specific KG.
comment: Accepted at IEEE ISDFS
☆ CoverUp: Coverage-Guided LLM-Based Test Generation
This paper presents CoverUp, a novel system that drives the generation of high-coverage Python regression tests via a combination of coverage analysis and large-language models (LLMs). CoverUp iteratively improves coverage, interleaving coverage analysis with dialogs with the LLM to focus its attention on as yet uncovered lines and branches. The resulting test suites significantly improve coverage over the current state of the art: compared to CodaMosa, a hybrid LLM / search-based software testing system, CoverUp substantially improves coverage across the board. On a per-module basis, CoverUp achieves median line coverage of 81% (vs. 62%), branch coverage of 53% (vs. 35%) and line+branch coverage of 78% (vs. 55%). We show that CoverUp's iterative, coverage-guided approach is crucial to its effectiveness, contributing to nearly half of its successes.
comment: 11 pages
☆ Frankenstein: Generating Semantic-Compositional 3D Scenes in One Tri-Plane
We present Frankenstein, a diffusion-based framework that can generate semantic-compositional 3D scenes in a single pass. Unlike existing methods that output a single, unified 3D shape, Frankenstein simultaneously generates multiple separated shapes, each corresponding to a semantically meaningful part. The 3D scene information is encoded in one single tri-plane tensor, from which multiple Singed Distance Function (SDF) fields can be decoded to represent the compositional shapes. During training, an auto-encoder compresses tri-planes into a latent space, and then the denoising diffusion process is employed to approximate the distribution of the compositional scenes. Frankenstein demonstrates promising results in generating room interiors as well as human avatars with automatically separated parts. The generated scenes facilitate many downstream applications, such as part-wise re-texturing, object rearrangement in the room or avatar cloth re-targeting.
comment: Video: https://youtu.be/lRn-HqyCrLI
Image Captioning in news report scenario
Image captioning strives to generate pertinent captions for specified images, situating itself at the crossroads of Computer Vision (CV) and Natural Language Processing (NLP). This endeavor is of paramount importance with far-reaching applications in recommendation systems, news outlets, social media, and beyond. Particularly within the realm of news reporting, captions are expected to encompass detailed information, such as the identities of celebrities captured in the images. However, much of the existing body of work primarily centers around understanding scenes and actions. In this paper, we explore the realm of image captioning specifically tailored for celebrity photographs, illustrating its broad potential for enhancing news industry practices. This exploration aims to augment automated news content generation, thereby facilitating a more nuanced dissemination of information. Our endeavor shows a broader horizon, enriching the narrative in news reporting through a more intuitive image captioning framework.
comment: 10 pages, 4 figures
Rumor Detection with a novel graph neural network approach
The wide spread of rumors on social media has caused a negative impact on people's daily life, leading to potential panic, fear, and mental health problems for the public. How to debunk rumors as early as possible remains a challenging problem. Existing studies mainly leverage information propagation structure to detect rumors, while very few works focus on correlation among users that they may coordinate to spread rumors in order to gain large popularity. In this paper, we propose a new detection model, that jointly learns both the representations of user correlation and information propagation to detect rumors on social media. Specifically, we leverage graph neural networks to learn the representations of user correlation from a bipartite graph that describes the correlations between users and source tweets, and the representations of information propagation with a tree structure. Then we combine the learned representations from these two modules to classify the rumors. Since malicious users intend to subvert our model after deployment, we further develop a greedy attack scheme to analyze the cost of three adversarial attacks: graph attack, comment attack, and joint attack. Evaluation results on two public datasets illustrate that the proposed MODEL outperforms the state-of-the-art rumor detection models. We also demonstrate our method performs well for early rumor detection. Moreover, the proposed detection method is more robust to adversarial attacks compared to the best existing method. Importantly, we show that it requires a high cost for attackers to subvert user correlation pattern, demonstrating the importance of considering user correlation for rumor detection.
comment: 10 pages, 5 figures
☆ Logic-based Explanations for Linear Support Vector Classifiers with Reject Option
Support Vector Classifier (SVC) is a well-known Machine Learning (ML) model for linear classification problems. It can be used in conjunction with a reject option strategy to reject instances that are hard to correctly classify and delegate them to a specialist. This further increases the confidence of the model. Given this, obtaining an explanation of the cause of rejection is important to not blindly trust the obtained results. While most of the related work has developed means to give such explanations for machine learning models, to the best of our knowledge none have done so for when reject option is present. We propose a logic-based approach with formal guarantees on the correctness and minimality of explanations for linear SVCs with reject option. We evaluate our approach by comparing it to Anchors, which is a heuristic algorithm for generating explanations. Obtained results show that our proposed method gives shorter explanations with reduced time cost.
comment: 16 pages, submitted to BRACIS 2023 (Brazilian Conference on Intelligent Systems), accepted version published in Intelligent Systems, LNCS, vol 14195
☆ Mixed-Initiative Human-Robot Teaming under Suboptimality with Online Bayesian Adaptation
For effective human-agent teaming, robots and other artificial intelligence (AI) agents must infer their human partner's abilities and behavioral response patterns and adapt accordingly. Most prior works make the unrealistic assumption that one or more teammates can act near-optimally. In real-world collaboration, humans and autonomous agents can be suboptimal, especially when each only has partial domain knowledge. In this work, we develop computational modeling and optimization techniques for enhancing the performance of suboptimal human-agent teams, where the human and the agent have asymmetric capabilities and act suboptimally due to incomplete environmental knowledge. We adopt an online Bayesian approach that enables a robot to infer people's willingness to comply with its assistance in a sequential decision-making game. Our user studies show that user preferences and team performance indeed vary with robot intervention styles, and our approach for mixed-initiative collaborations enhances objective team performance ($p<.001$) and subjective measures, such as user's trust ($p<.001$) and perceived likeability of the robot ($p<.001$).
comment: 8 pages, 4 pages for supplementary
☆ An Analytic Solution to Covariance Propagation in Neural Networks AISTATS 2024
Uncertainty quantification of neural networks is critical to measuring the reliability and robustness of deep learning systems. However, this often involves costly or inaccurate sampling methods and approximations. This paper presents a sample-free moment propagation technique that propagates mean vectors and covariance matrices across a network to accurately characterize the input-output distributions of neural networks. A key enabler of our technique is an analytic solution for the covariance of random variables passed through nonlinear activation functions, such as Heaviside, ReLU, and GELU. The wide applicability and merits of the proposed technique are shown in experiments analyzing the input-output distributions of trained neural networks and training Bayesian neural networks.
comment: Accepted to AISTATS 2024
☆ Multi-Task Learning with Multi-Task Optimization
Multi-task learning solves multiple correlated tasks. However, conflicts may exist between them. In such circumstances, a single solution can rarely optimize all the tasks, leading to performance trade-offs. To arrive at a set of optimized yet well-distributed models that collectively embody different trade-offs in one algorithmic pass, this paper proposes to view Pareto multi-task learning through the lens of multi-task optimization. Multi-task learning is first cast as a multi-objective optimization problem, which is then decomposed into a diverse set of unconstrained scalar-valued subproblems. These subproblems are solved jointly using a novel multi-task gradient descent method, whose uniqueness lies in the iterative transfer of model parameters among the subproblems during the course of optimization. A theorem proving faster convergence through the inclusion of such transfers is presented. We investigate the proposed multi-task learning with multi-task optimization for solving various problem settings including image classification, scene understanding, and multi-target regression. Comprehensive experiments confirm that the proposed method significantly advances the state-of-the-art in discovering sets of Pareto-optimized models. Notably, on the large image dataset we tested on, namely NYUv2, the hypervolume convergence achieved by our method was found to be nearly two times faster than the next-best among the state-of-the-art.
☆ One Masked Model is All You Need for Sensor Fault Detection, Isolation and Accommodation IJCNN 2024
Accurate and reliable sensor measurements are critical for ensuring the safety and longevity of complex engineering systems such as wind turbines. In this paper, we propose a novel framework for sensor fault detection, isolation, and accommodation (FDIA) using masked models and self-supervised learning. Our proposed approach is a general time series modeling approach that can be applied to any neural network (NN) model capable of sequence modeling, and captures the complex spatio-temporal relationships among different sensors. During training, the proposed masked approach creates a random mask, which acts like a fault, for one or more sensors, making the training and inference task unified: finding the faulty sensors and correcting them. We validate our proposed technique on both a public dataset and a real-world dataset from GE offshore wind turbines, and demonstrate its effectiveness in detecting, diagnosing and correcting sensor faults. The masked model not only simplifies the overall FDIA pipeline, but also outperforms existing approaches. Our proposed technique has the potential to significantly improve the accuracy and reliability of sensor measurements in complex engineering systems in real-time, and could be applied to other types of sensors and engineering systems in the future. We believe that our proposed framework can contribute to the development of more efficient and effective FDIA techniques for a wide range of applications.
comment: Accepted by the 2024 International Joint Conference on Neural Networks (IJCNN 2024)
☆ A Survey on Consumer IoT Traffic: Security and Privacy
For the past few years, the Consumer Internet of Things (CIoT) has entered public lives. While CIoT has improved the convenience of people's daily lives, it has also brought new security and privacy concerns. In this survey, we try to figure out what researchers can learn about the security and privacy of CIoT by traffic analysis, a popular method in the security community. From the security and privacy perspective, this survey seeks out the new characteristics in CIoT traffic analysis, the state-of-the-art progress in CIoT traffic analysis, and the challenges yet to be solved. We collected 310 papers from January 2018 to December 2023 related to CIoT traffic analysis from the security and privacy perspective and summarized the process of CIoT traffic analysis in which the new characteristics of CIoT are identified. Then, we detail existing works based on five application goals: device fingerprinting, user activity inference, malicious traffic analysis, security analysis, and measurement. At last, we discuss the new challenges and future research directions.
☆ What Happens to a Dataset Transformed by a Projection-based Concept Removal Method?
We investigate the behavior of methods that use linear projections to remove information about a concept from a language representation, and we consider the question of what happens to a dataset transformed by such a method. A theoretical analysis and experiments on real-world and synthetic data show that these methods inject strong statistical dependencies into the transformed datasets. After applying such a method, the representation space is highly structured: in the transformed space, an instance tends to be located near instances of the opposite label. As a consequence, the original labeling can in some cases be reconstructed by applying an anti-clustering method.
☆ Complementary Recommendation in E-commerce: Definition, Approaches, and Future Directions
In recent years, complementary recommendation has received extensive attention in the e-commerce domain. In this paper, we comprehensively summarize and compare 34 representative studies conducted between 2009 and 2024. Firstly, we compare the data and methods used for modeling complementary relationships between products, including simple complementarity and more complex scenarios such as asymmetric complementarity, the coexistence of substitution and complementarity relationships between products, and varying degrees of complementarity between different pairs of products. Next, we classify and compare the models based on the research problems of complementary recommendation, such as diversity, personalization, and cold-start. Furthermore, we provide a comparative analysis of experimental results from different studies conducted on the same dataset, which helps identify the strengths and weaknesses of the research. Compared to previous surveys, this paper provides a more updated and comprehensive summary of the research, discusses future research directions, and contributes to the advancement of this field.
comment: 20 pages,9 figures
☆ SSHPool: The Separated Subgraph-based Hierarchical Pooling
In this paper, we develop a novel local graph pooling method, namely the Separated Subgraph-based Hierarchical Pooling (SSHPool), for graph classification. To this end, we commence by assigning the nodes of a sample graph into different clusters, resulting in a family of separated subgraphs. We individually employ a local graph convolution units as the local structure to further compress each subgraph into a coarsened node, transforming the original graph into a coarsened graph. Since these subgraphs are separated by different clusters and the structural information cannot be propagated between them, the local convolution operation can significantly avoid the over-smoothing problem arising in most existing Graph Neural Networks (GNNs). By hierarchically performing the proposed procedures on the resulting coarsened graph, the proposed SSHPool can effectively extract the hierarchical global feature of the original graph structure, encapsulating rich intrinsic structural characteristics. Furthermore, we develop an end-to-end GNN framework associated with the proposed SSHPool module for graph classification. Experimental results demonstrate the superior performance of the proposed model on real-world datasets, significantly outperforming state-of-the-art GNN methods in terms of the classification accuracies.
☆ AKBR: Learning Adaptive Kernel-based Representations for Graph Classification
In this paper, we propose a new model to learn Adaptive Kernel-based Representations (AKBR) for graph classification. Unlike state-of-the-art R-convolution graph kernels that are defined by merely counting any pair of isomorphic substructures between graphs and cannot provide an end-to-end learning mechanism for the classifier, the proposed AKBR approach aims to define an end-to-end representation learning model to construct an adaptive kernel matrix for graphs. To this end, we commence by leveraging a novel feature-channel attention mechanism to capture the interdependencies between different substructure invariants of original graphs. The proposed AKBR model can thus effectively identify the structural importance of different substructures, and compute the R-convolution kernel between pairwise graphs associated with the more significant substructures specified by their structural attentions. Since each row of the resulting kernel matrix can be theoretically seen as the embedding vector of a sample graph, the proposed AKBR model is able to directly employ the resulting kernel matrix as the graph feature matrix and input it into the classifier for classification (i.e., the SoftMax layer), naturally providing an end-to-end learning architecture between the kernel computation as well as the classifier. Experimental results show that the proposed AKBR model outperforms existing state-of-the-art graph kernels and deep learning methods on standard graph benchmarks.
☆ WangchanLion and WangchanX MRC Eval
This technical report describes the development of WangchanLion, an instruction fine-tuned model focusing on Machine Reading Comprehension (MRC) in the Thai language. Our model is based on SEA-LION and a collection of instruction following datasets. To promote open research and reproducibility, we publically release all training data, code, and the final model weights under the Apache-2 license. To assess the contextual understanding capability, we conducted extensive experimental studies using two Thai MRC datasets, XQuAD and Iapp_wiki_qa_squad. Experimental results demonstrate the model's ability to comprehend the context and produce an answer faithful to the reference one in 0-shot and 1-shot settings. In addition, our evaluation goes beyond the traditional MRC. We propose a new evaluation scheme assessing the answer's correctness, helpfulness, conciseness, and contextuality. Evaluation results provide insight into how we can improve our model in the future. Our code is public at https://github.com/vistec-AI/WangchanLion.
Self-Supervised Multi-Frame Neural Scene Flow
Neural Scene Flow Prior (NSFP) and Fast Neural Scene Flow (FNSF) have shown remarkable adaptability in the context of large out-of-distribution autonomous driving. Despite their success, the underlying reasons for their astonishing generalization capabilities remain unclear. Our research addresses this gap by examining the generalization capabilities of NSFP through the lens of uniform stability, revealing that its performance is inversely proportional to the number of input point clouds. This finding sheds light on NSFP's effectiveness in handling large-scale point cloud scene flow estimation tasks. Motivated by such theoretical insights, we further explore the improvement of scene flow estimation by leveraging historical point clouds across multiple frames, which inherently increases the number of point clouds. Consequently, we propose a simple and effective method for multi-frame point cloud scene flow estimation, along with a theoretical evaluation of its generalization abilities. Our analysis confirms that the proposed method maintains a limited generalization error, suggesting that adding multiple frames to the scene flow optimization process does not detract from its generalizability. Extensive experimental results on large-scale autonomous driving Waymo Open and Argoverse lidar datasets demonstrate that the proposed method achieves state-of-the-art performance.
☆ Opportunities and challenges in the application of large artificial intelligence models in radiology
Influenced by ChatGPT, artificial intelligence (AI) large models have witnessed a global upsurge in large model research and development. As people enjoy the convenience by this AI large model, more and more large models in subdivided fields are gradually being proposed, especially large models in radiology imaging field. This article first introduces the development history of large models, technical details, workflow, working principles of multimodal large models and working principles of video generation large models. Secondly, we summarize the latest research progress of AI large models in radiology education, radiology report generation, applications of unimodal and multimodal radiology. Finally, this paper also summarizes some of the challenges of large AI models in radiology, with the aim of better promoting the rapid revolution in the field of radiography.
☆ A Transformer approach for Electricity Price Forecasting
This paper presents a novel approach to electricity price forecasting (EPF) using a pure Transformer model. As opposed to other alternatives, no other recurrent network is used in combination to the attention mechanism. Hence, showing that the attention layer is enough for capturing the temporal patterns. The paper also provides fair comparison of the models using the open-source EPF toolbox and provide the code to enhance reproducibility and transparency in EPF research. The results show that the Transformer model outperforms traditional methods, offering a promising solution for reliable and sustainable power system operation.
comment: 7 pages
☆ Evaluating Fairness Metrics Across Borders from Human Perceptions
Which fairness metrics are appropriately applicable in your contexts? There may be instances of discordance regarding the perception of fairness, even when the outcomes comply with established fairness metrics. Several surveys have been conducted to evaluate fairness metrics with human perceptions of fairness. However, these surveys were limited in scope, including only a few hundred participants within a single country. In this study, we conduct an international survey to evaluate the appropriateness of various fairness metrics in decision-making scenarios. We collected responses from 1,000 participants in each of China, France, Japan, and the United States, amassing a total of 4,000 responses, to analyze the preferences of fairness metrics. Our survey consists of three distinct scenarios paired with four fairness metrics, and each participant answers their preference for the fairness metric in each case. This investigation explores the relationship between personal attributes and the choice of fairness metrics, uncovering a significant influence of national context on these preferences.
☆ Specifying Agent Ethics (Blue Sky Ideas)
We consider the question of what properties a Machine Ethics system should have. This question is complicated by the existence of ethical dilemmas with no agreed upon solution. We provide an example to motivate why we do not believe falling back on the elicitation of values from stakeholders is sufficient to guarantee correctness of such systems. We go on to define two broad categories of ethical property that have arisen in our own work and present a challenge to the community to approach this question in a more systematic way.
comment: To appear in Coordination, Organizations, Institutions, Norms and Ethics for Governance of Multi-Agent Systems 2024
☆ Can Language Models Pretend Solvers? Logic Code Simulation with LLMs
Transformer-based large language models (LLMs) have demonstrated significant potential in addressing logic problems. capitalizing on the great capabilities of LLMs for code-related activities, several frameworks leveraging logical solvers for logic reasoning have been proposed recently. While existing research predominantly focuses on viewing LLMs as natural language logic solvers or translators, their roles as logic code interpreters and executors have received limited attention. This study delves into a novel aspect, namely logic code simulation, which forces LLMs to emulate logical solvers in predicting the results of logical programs. To further investigate this novel task, we formulate our three research questions: Can LLMs efficiently simulate the outputs of logic codes? What strength arises along with logic code simulation? And what pitfalls? To address these inquiries, we curate three novel datasets tailored for the logic code simulation task and undertake thorough experiments to establish the baseline performance of LLMs in code simulation. Subsequently, we introduce a pioneering LLM-based code simulation technique, Dual Chains of Logic (DCoL). This technique advocates a dual-path thinking approach for LLMs, which has demonstrated state-of-the-art performance compared to other LLM prompt strategies, achieving a notable improvement in accuracy by 7.06% with GPT-4-Turbo.
comment: 12 pages, 8 figures
☆ The Interplay of Learning, Analytics, and Artificial Intelligence in Education
This paper presents a multi dimensional view of AI's role in learning and education, emphasizing the intricate interplay between AI, analytics, and the learning processes. Here, I challenge the prevalent narrow conceptualization of AI as stochastic tools, as exemplified in generative AI, and argue for the importance of alternative conceptualisations of AI. I highlight the differences between human intelligence and artificial information processing, the cognitive diversity inherent in AI algorithms, and posit that AI can also serve as an instrument for understanding human learning. Early learning sciences and AI in Education research, which saw AI as an analogy for human intelligence, have diverged from this perspective, prompting a need to rekindle this connection. The paper presents three unique conceptualizations of AI in education: the externalization of human cognition, the internalization of AI models to influence human thought processes, and the extension of human cognition via tightly integrated human-AI systems. Examples from current research and practice are examined as instances of the three conceptualisations, highlighting the potential value and limitations of each conceptualisation for education, as well as the perils of overemphasis on externalising human cognition as exemplified in today's hype surrounding generative AI tools. The paper concludes with an advocacy for a broader educational approach that includes educating people about AI and innovating educational systems to remain relevant in an AI enabled world.
comment: 20 pages, 7 figures, this paper is based on the keynote talk given by the author at the ACM International Conference on Learning Analytics & Knowledge (LAK) 2024 in Kyoto, Japan. https://www.solaresearch.org/events/lak/lak24/keynotes/
☆ Landmark-Guided Cross-Speaker Lip Reading with Mutual Information Regularization LREC
Lip reading, the process of interpreting silent speech from visual lip movements, has gained rising attention for its wide range of realistic applications. Deep learning approaches greatly improve current lip reading systems. However, lip reading in cross-speaker scenarios where the speaker identity changes, poses a challenging problem due to inter-speaker variability. A well-trained lip reading system may perform poorly when handling a brand new speaker. To learn a speaker-robust lip reading model, a key insight is to reduce visual variations across speakers, avoiding the model overfitting to specific speakers. In this work, in view of both input visual clues and latent representations based on a hybrid CTC/attention architecture, we propose to exploit the lip landmark-guided fine-grained visual clues instead of frequently-used mouth-cropped images as input features, diminishing speaker-specific appearance characteristics. Furthermore, a max-min mutual information regularization approach is proposed to capture speaker-insensitive latent representations. Experimental evaluations on public lip reading datasets demonstrate the effectiveness of the proposed approach under the intra-speaker and inter-speaker conditions.
comment: To appear in LREC-COLING 2024
☆ Robust Diffusion Models for Adversarial Purification
Diffusion models (DMs) based adversarial purification (AP) has shown to be the most powerful alternative to adversarial training (AT). However, these methods neglect the fact that pre-trained diffusion models themselves are not robust to adversarial attacks as well. Additionally, the diffusion process can easily destroy semantic information and generate a high quality image but totally different from the original input image after the reverse process, leading to degraded standard accuracy. To overcome these issues, a natural idea is to harness adversarial training strategy to retrain or fine-tune the pre-trained diffusion model, which is computationally prohibitive. We propose a novel robust reverse process with adversarial guidance, which is independent of given pre-trained DMs and avoids retraining or fine-tuning the DMs. This robust guidance can not only ensure to generate purified examples retaining more semantic content but also mitigate the accuracy-robustness trade-off of DMs for the first time, which also provides DM-based AP an efficient adaptive ability to new attacks. Extensive experiments are conducted to demonstrate that our method achieves the state-of-the-art results and exhibits generalization against different attacks.
☆ A Temporal Graph Network Framework for Dynamic Recommendation AAAI 2024
Recommender systems, crucial for user engagement on platforms like e-commerce and streaming services, often lag behind users' evolving preferences due to static data reliance. After Temporal Graph Networks (TGNs) were proposed, various studies have shown that TGN can significantly improve situations where the features of nodes and edges dynamically change over time. However, despite its promising capabilities, it has not been directly applied in recommender systems to date. Our study bridges this gap by directly implementing Temporal Graph Networks (TGN) in recommender systems, a first in this field. Using real-world datasets and a range of graph and history embedding methods, we show TGN's adaptability, confirming its effectiveness in dynamic recommendation scenarios.
comment: Presented at the AAAI 2024 Workshop on Recommendation Ecosystems: Modeling, Optimization and Incentive Design
☆ Qibo: A Large Language Model for Traditional Chinese Medicine
In the field of Artificial Intelligence, Large Language Models (LLMs) have demonstrated significant advances in user intent understanding and response in a number of specialized domains, including medicine, law, and finance. However, in the unique domain of traditional Chinese medicine (TCM), the performance enhancement of LLMs is challenged by the essential differences between its theories and modern medicine, as well as the lack of specialized corpus resources. In this paper, we aim to construct and organize a professional corpus in the field of TCM, to endow the large model with professional knowledge that is characteristic of TCM theory, and to successfully develop the Qibo model based on LLaMA, which is the first LLM in the field of TCM to undergo a complete training process from pre-training to Supervised Fine-Tuning (SFT). Furthermore, we develop the Qibo-benchmark, a specialized tool for evaluating the performance of LLMs, which is a specialized tool for evaluating the performance of LLMs in the TCM domain. This tool will provide an important basis for quantifying and comparing the understanding and application capabilities of different models in the field of traditional Chinese medicine, and provide guidance for future research directions and practical applications of intelligent assistants for traditional Chinese medicine. Finally, we conducted sufficient experiments to prove that Qibo has good performance in the field of traditional Chinese medicine.
☆ Semantic Is Enough: Only Semantic Information For NeRF Reconstruction
Recent research that combines implicit 3D representation with semantic information, like Semantic-NeRF, has proven that NeRF model could perform excellently in rendering 3D structures with semantic labels. This research aims to extend the Semantic Neural Radiance Fields (Semantic-NeRF) model by focusing solely on semantic output and removing the RGB output component. We reformulate the model and its training procedure to leverage only the cross-entropy loss between the model semantic output and the ground truth semantic images, removing the colour data traditionally used in the original Semantic-NeRF approach. We then conduct a series of identical experiments using the original and the modified Semantic-NeRF model. Our primary objective is to obverse the impact of this modification on the model performance by Semantic-NeRF, focusing on tasks such as scene understanding, object detection, and segmentation. The results offer valuable insights into the new way of rendering the scenes and provide an avenue for further research and development in semantic-focused 3D scene understanding.
☆ RPMArt: Towards Robust Perception and Manipulation for Articulated Objects IROS 2024
Articulated objects are commonly found in daily life. It is essential that robots can exhibit robust perception and manipulation skills for articulated objects in real-world robotic applications. However, existing methods for articulated objects insufficiently address noise in point clouds and struggle to bridge the gap between simulation and reality, thus limiting the practical deployment in real-world scenarios. To tackle these challenges, we propose a framework towards Robust Perception and Manipulation for Articulated Objects (RPMArt), which learns to estimate the articulation parameters and manipulate the articulation part from the noisy point cloud. Our primary contribution is a Robust Articulation Network (RoArtNet) that is able to predict both joint parameters and affordable points robustly by local feature learning and point tuple voting. Moreover, we introduce an articulation-aware classification scheme to enhance its ability for sim-to-real transfer. Finally, with the estimated affordable point and articulation joint constraint, the robot can generate robust actions to manipulate articulated objects. After learning only from synthetic data, RPMArt is able to transfer zero-shot to real-world articulated objects. Experimental results confirm our approach's effectiveness, with our framework achieving state-of-the-art performance in both noise-added simulation and real-world environments. The code and data will be open-sourced for reproduction. More results are published on the project website at https://r-pmart.github.io .
comment: 8 pages, 7 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024), project website at https://r-pmart.github.io
☆ Fill in the ____ (a Diffusion-based Image Inpainting Pipeline)
Image inpainting is the process of taking an image and generating lost or intentionally occluded portions. Inpainting has countless applications including restoring previously damaged pictures, restoring the quality of images that have been degraded due to compression, and removing unwanted objects/text. Modern inpainting techniques have shown remarkable ability in generating sensible completions for images with mask occlusions. In our paper, an overview of the progress of inpainting techniques will be provided, along with identifying current leading approaches, focusing on their strengths and weaknesses. A critical gap in these existing models will be addressed, focusing on the ability to prompt and control what exactly is generated. We will additionally justify why we think this is the natural next progressive step that inpainting models must take, and provide multiple approaches to implementing this functionality. Finally, we will evaluate the results of our approaches by qualitatively checking whether they generate high-quality images that correctly inpaint regions with the objects that they are instructed to produce.
☆ A Federated Parameter Aggregation Method for Node Classification Tasks with Different Graph Network Structures
Over the past few years, federated learning has become widely used in various classical machine learning fields because of its collaborative ability to train data from multiple sources without compromising privacy. However, in the area of graph neural networks, the nodes and network structures of graphs held by clients are different in many practical applications, and the aggregation method that directly shares model gradients cannot be directly applied to this scenario. Therefore, this work proposes a federated aggregation method FLGNN applied to various graph federation scenarios and investigates the aggregation effect of parameter sharing at each layer of the graph neural network model. The effectiveness of the federated aggregation method FLGNN is verified by experiments on real datasets. Additionally, for the privacy security of FLGNN, this paper designs membership inference attack experiments and differential privacy defense experiments. The results show that FLGNN performs good robustness, and the success rate of privacy theft is further reduced by adding differential privacy defense methods.
☆ Diverse Representation Embedding for Lifelong Person Re-Identification
Lifelong Person Re-Identification (LReID) aims to continuously learn from successive data streams, matching individuals across multiple cameras. The key challenge for LReID is how to effectively preserve old knowledge while learning new information incrementally. Task-level domain gaps and limited old task datasets are key factors leading to catastrophic forgetting in ReLD, which are overlooked in existing methods. To alleviate this problem, we propose a novel Diverse Representation Embedding (DRE) framework for LReID. The proposed DRE preserves old knowledge while adapting to new information based on instance-level and task-level layout. Concretely, an Adaptive Constraint Module (ACM) is proposed to implement integration and push away operations between multiple representations, obtaining dense embedding subspace for each instance to improve matching ability on limited old task datasets. Based on the processed diverse representation, we interact knowledge between the adjustment model and the learner model through Knowledge Update (KU) and Knowledge Preservation (KP) strategies at the task-level layout, which reduce the task-wise domain gap on both old and new tasks, and exploit diverse representation of each instance in limited datasets from old tasks, improving model performance for extended periods. Extensive experiments were conducted on eleven Re-ID datasets, including five seen datasets for training in order-1 and order-2 orders and six unseen datasets for inference. Compared to state-of-the-art methods, our method achieves significantly improved performance in holistic, large-scale, and occluded datasets.
comment: 11 pages,7 Tables,3 Figures
☆ Multi-Scale Spatio-Temporal Graph Convolutional Network for Facial Expression Spotting
Facial expression spotting is a significant but challenging task in facial expression analysis. The accuracy of expression spotting is affected not only by irrelevant facial movements but also by the difficulty of perceiving subtle motions in micro-expressions. In this paper, we propose a Multi-Scale Spatio-Temporal Graph Convolutional Network (SpoT-GCN) for facial expression spotting. To extract more robust motion features, we track both short- and long-term motion of facial muscles in compact sliding windows whose window length adapts to the temporal receptive field of the network. This strategy, termed the receptive field adaptive sliding window strategy, effectively magnifies the motion features while alleviating the problem of severe head movement. The subtle motion features are then converted to a facial graph representation, whose spatio-temporal graph patterns are learned by a graph convolutional network. This network learns both local and global features from multiple scales of facial graph structures using our proposed facial local graph pooling (FLGP). Furthermore, we introduce supervised contrastive learning to enhance the discriminative capability of our model for difficult-to-classify frames. The experimental results on the SAMM-LV and CAS(ME)^2 datasets demonstrate that our method achieves state-of-the-art performance, particularly in micro-expression spotting. Ablation studies further verify the effectiveness of our proposed modules.
comment: Accepted by FG2024
☆ Knowledge-guided Machine Learning: Current Trends and Future Prospects
This paper presents an overview of scientific modeling and discusses the complementary strengths and weaknesses of ML methods for scientific modeling in comparison to process-based models. It also provides an introduction to the current state of research in the emerging field of scientific knowledge-guided machine learning (KGML) that aims to use both scientific knowledge and data in ML frameworks to achieve better generalizability, scientific consistency, and explainability of results. We discuss different facets of KGML research in terms of the type of scientific knowledge used, the form of knowledge-ML integration explored, and the method for incorporating scientific knowledge in ML. We also discuss some of the common categories of use cases in environmental sciences where KGML methods are being developed, using illustrative examples in each category.
☆ Towards Two-Stream Foveation-based Active Vision Learning
Deep neural network (DNN) based machine perception frameworks process the entire input in a one-shot manner to provide answers to both "what object is being observed" and "where it is located". In contrast, the "two-stream hypothesis" from neuroscience explains the neural processing in the human visual cortex as an active vision system that utilizes two separate regions of the brain to answer the what and the where questions. In this work, we propose a machine learning framework inspired by the "two-stream hypothesis" and explore the potential benefits that it offers. Specifically, the proposed framework models the following mechanisms: 1) ventral (what) stream focusing on the input regions perceived by the fovea part of an eye (foveation), 2) dorsal (where) stream providing visual guidance, and 3) iterative processing of the two streams to calibrate visual focus and process the sequence of focused image patches. The training of the proposed framework is accomplished by label-based DNN training for the ventral stream model and reinforcement learning for the dorsal stream model. We show that the two-stream foveation-based learning is applicable to the challenging task of weakly-supervised object localization (WSOL), where the training data is limited to the object class or its attributes. The framework is capable of both predicting the properties of an object and successfully localizing it by predicting its bounding box. We also show that, due to the independent nature of the two streams, the dorsal model can be applied on its own to unseen images to localize objects from different datasets.
comment: 18 pages, 14 figures, Under consideration at IEEE Transactions on Cognitive and Developmental Systems
☆ CBGT-Net: A Neuromimetic Architecture for Robust Classification of Streaming Data
This paper describes CBGT-Net, a neural network model inspired by the cortico-basal ganglia-thalamic (CBGT) circuits found in mammalian brains. Unlike traditional neural network models, which either generate an output for each provided input, or an output after a fixed sequence of inputs, the CBGT-Net learns to produce an output after a sufficient criteria for evidence is achieved from a stream of observed data. For each observation, the CBGT-Net generates a vector that explicitly represents the amount of evidence the observation provides for each potential decision, accumulates the evidence over time, and generates a decision when the accumulated evidence exceeds a pre-defined threshold. We evaluate the proposed model on two image classification tasks, where models need to predict image categories based on a stream of small patches extracted from the image. We show that the CBGT-Net provides improved accuracy and robustness compared to models trained to classify from a single patch, and models leveraging an LSTM layer to classify from a fixed sequence length of patches.
♻ ☆ Make a Donut: Hierarchical EMD-Space Planning for Zero-Shot Deformable Manipulation with Tools
Deformable object manipulation stands as one of the most captivating yet formidable challenges in robotics. While previous techniques have predominantly relied on learning latent dynamics through demonstrations, typically represented as either particles or images, there exists a pertinent limitation: acquiring suitable demonstrations, especially for long-horizon tasks, can be elusive. Moreover, basing learning entirely on demonstrations can hamper the model's ability to generalize beyond the demonstrated tasks. In this work, we introduce a demonstration-free hierarchical planning approach capable of tackling intricate long-horizon tasks without necessitating any training. We employ large language models (LLMs) to articulate a high-level, stage-by-stage plan corresponding to a specified task. For every individual stage, the LLM provides both the tool's name and the Python code to craft intermediate subgoal point clouds. With the tool and subgoal for a particular stage at our disposal, we present a granular closed-loop model predictive control strategy. This leverages Differentiable Physics with Point-to-Point correspondence (DiffPhysics-P2P) loss in the earth mover distance (EMD) space, applied iteratively. Experimental findings affirm that our technique surpasses multiple benchmarks in dough manipulation, spanning both short and long horizons. Remarkably, our model demonstrates robust generalization capabilities to novel and previously unencountered complex tasks without any preliminary demonstrations. We further substantiate our approach with experimental trials on real-world robotic platforms. Our project page: https://qq456cvb.github.io/projects/donut.
comment: 8 pages
♻ ☆ VQPy: An Object-Oriented Approach to Modern Video Analytics
Video analytics is widely used in contemporary systems and services. At the forefront of video analytics are video queries that users develop to find objects of particular interest. Building upon the insight that video objects (e.g., human, animals, cars, etc.), the center of video analytics, are similar in spirit to objects modeled by traditional object-oriented languages, we propose to develop an object-oriented approach to video analytics. This approach, named VQPy, consists of a frontend$\unicode{x2015}$a Python variant with constructs that make it easy for users to express video objects and their interactions$\unicode{x2015}$as well as an extensible backend that can automatically construct and optimize pipelines based on video objects. We have implemented and open-sourced VQPy, which has been productized in Cisco as part of its DeepVision framework.
comment: MLSys'24
♻ ☆ Unmasking and Improving Data Credibility: A Study with Datasets for Training Harmless Language Models ICLR 2024
Language models have shown promise in various tasks but can be affected by undesired data during training, fine-tuning, or alignment. For example, if some unsafe conversations are wrongly annotated as safe ones, the model fine-tuned on these samples may be harmful. Therefore, the correctness of annotations, i.e., the credibility of the dataset, is important. This study focuses on the credibility of real-world datasets, including the popular benchmarks Jigsaw Civil Comments, Anthropic Harmless & Red Team, PKU BeaverTails & SafeRLHF, that can be used for training a harmless language model. Given the cost and difficulty of cleaning these datasets by humans, we introduce a systematic framework for evaluating the credibility of datasets, identifying label errors, and evaluating the influence of noisy labels in the curated language data, specifically focusing on unsafe comments and conversation classification. With the framework, we find and fix an average of 6.16% label errors in 11 datasets constructed from the above benchmarks. The data credibility and downstream learning performance can be remarkably improved by directly fixing label errors, indicating the significance of cleaning existing real-world datasets. We provide an open-source tool, Docta, for data cleaning at https://github.com/Docta-ai/docta.
comment: ICLR 2024
♻ ☆ Latent Dataset Distillation with Diffusion Models
The efficacy of machine learning has traditionally relied on the availability of increasingly larger datasets. However, large datasets pose storage challenges and contain non-influential samples, which could be ignored during training without impacting the final accuracy of the model. In response to these limitations, the concept of distilling the information on a dataset into a condensed set of (synthetic) samples, namely a distilled dataset, emerged. One crucial aspect is the selected architecture (usually ConvNet) for linking the original and synthetic datasets. However, the final accuracy is lower if the employed model architecture differs from the model used during distillation. Another challenge is the generation of high-resolution images, e.g., 128x128 and higher. In this paper, we propose Latent Dataset Distillation with Diffusion Models (LD3M) that combine diffusion in latent space with dataset distillation to tackle both challenges. LD3M incorporates a novel diffusion process tailored for dataset distillation, which improves the gradient norms for learning synthetic images. By adjusting the number of diffusion steps, LD3M also offers a straightforward way of controlling the trade-off between speed and accuracy. We evaluate our approach in several ImageNet subsets and for high-resolution images (128x128 and 256x256). As a result, LD3M consistently outperforms state-of-the-art distillation techniques by up to 4.8 p.p. and 4.2 p.p. for 1 and 10 images per class, respectively.
♻ ☆ Detection of diabetic retinopathy using longitudinal self-supervised learning MICCAI
Longitudinal imaging is able to capture both static anatomical structures and dynamic changes in disease progression towards earlier and better patient-specific pathology management. However, conventional approaches for detecting diabetic retinopathy (DR) rarely take advantage of longitudinal information to improve DR analysis. In this work, we investigate the benefit of exploiting self-supervised learning with a longitudinal nature for DR diagnosis purposes. We compare different longitudinal self-supervised learning (LSSL) methods to model the disease progression from longitudinal retinal color fundus photographs (CFP) to detect early DR severity changes using a pair of consecutive exams. The experiments were conducted on a longitudinal DR screening dataset with or without those trained encoders (LSSL) acting as a longitudinal pretext task. Results achieve an AUC of 0.875 for the baseline (model trained from scratch) and an AUC of 0.96 (95% CI: 0.9593-0.9655 DeLong test) with a p-value < 2.2e-16 on early fusion using a simple ResNet alike architecture with frozen LSSL weights, suggesting that the LSSL latent space enables to encode the dynamic of DR progression.
comment: Accepted preprint for presentation at MICCAI-OMIA
♻ ☆ Lemur: Integrating Large Language Models in Automated Program Verification ICLR'24
The demonstrated code-understanding capability of LLMs raises the question of whether they can be used for automated program verification, a task that demands high-level abstract reasoning about program properties that is challenging for verification tools. We propose a general methodology to combine the power of LLMs and automated reasoners for automated program verification. We formally describe this methodology as a set of derivation rules and prove its soundness. We instantiate the calculus as a sound automated verification procedure, which led to practical improvements on a set of synthetic and competition benchmarks.
comment: Accepted at ICLR'24
♻ ☆ C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersion ICLR 2024
In deep learning, test-time adaptation has gained attention as a method for model fine-tuning without the need for labeled data. A prime exemplification is the recently proposed test-time prompt tuning for large-scale vision-language models such as CLIP. Unfortunately, these prompts have been mainly developed to improve accuracy, overlooking the importance of calibration, which is a crucial aspect for quantifying prediction uncertainty. However, traditional calibration methods rely on substantial amounts of labeled data, making them impractical for test-time scenarios. To this end, this paper explores calibration during test-time prompt tuning by leveraging the inherent properties of CLIP. Through a series of observations, we find that the prompt choice significantly affects the calibration in CLIP, where the prompts leading to higher text feature dispersion result in better-calibrated predictions. Introducing the Average Text Feature Dispersion (ATFD), we establish its relationship with calibration error and present a novel method, Calibrated Test-time Prompt Tuning (C-TPT), for optimizing prompts during test-time with enhanced calibration. Through extensive experiments on different CLIP architectures and datasets, we show that C-TPT can effectively improve the calibration of test-time prompt tuning without needing labeled data. The code is publicly accessible at https://github.com/hee-suk-yoon/C-TPT.
comment: ICLR 2024
♻ ☆ Our Model Achieves Excellent Performance on MovieLens: What Does it Mean?
A typical benchmark dataset for recommender system (RecSys) evaluation consists of user-item interactions generated on a platform within a time period. The interaction generation mechanism partially explains why a user interacts with (e.g., like, purchase, rate) an item, and the context of when a particular interaction happened. In this study, we conduct a meticulous analysis of the MovieLens dataset and explain the potential impact of using the dataset for evaluating recommendation algorithms. We make a few main findings from our analysis. First, there are significant differences in user interactions at the different stages when a user interacts with the MovieLens platform. The early interactions largely define the user portrait which affects the subsequent interactions. Second, user interactions are highly affected by the candidate movies that are recommended by the platform's internal recommendation algorithm(s). Third, changing the order of user interactions makes it more difficult for sequential algorithms to capture the progressive interaction process. We further discuss the discrepancy between the interaction generation mechanism that is employed by the MovieLens system and that of typical real-world recommendation scenarios. In summary, the MovieLens platform demonstrates an efficient and effective way of collecting user preferences to address cold-starts. However, models that achieve excellent recommendation accuracy on the MovieLens dataset may not demonstrate superior performance in practice, for at least two kinds of differences: (i) the differences in the contexts of user-item interaction generation, and (ii) the differences in user knowledge about the item collections. While results on MovieLens can be useful as a reference, they should not be solely relied upon as the primary justification for the effectiveness of a recommendation system model.
♻ ☆ Rethinking Channel Dependence for Multivariate Time Series Forecasting: Learning from Leading Indicators ICLR 2024
Recently, channel-independent methods have achieved state-of-the-art performance in multivariate time series (MTS) forecasting. Despite reducing overfitting risks, these methods miss potential opportunities in utilizing channel dependence for accurate predictions. We argue that there exist locally stationary lead-lag relationships between variates, i.e., some lagged variates may follow the leading indicators within a short time period. Exploiting such channel dependence is beneficial since leading indicators offer advance information that can be used to reduce the forecasting difficulty of the lagged variates. In this paper, we propose a new method named LIFT that first efficiently estimates leading indicators and their leading steps at each time step and then judiciously allows the lagged variates to utilize the advance information from leading indicators. LIFT plays as a plugin that can be seamlessly collaborated with arbitrary time series forecasting methods. Extensive experiments on six real-world datasets demonstrate that LIFT improves the state-of-the-art methods by 5.5% in average forecasting performance. Our code is available at https://github.com/SJTU-Quant/LIFT.
comment: Accepted to ICLR 2024
♻ ☆ Less is More: A Closer Look at Semantic-based Few-Shot Learning
Few-shot Learning aims to learn and distinguish new categories with a very limited number of available images, presenting a significant challenge in the realm of deep learning. Recent researchers have sought to leverage the additional textual or linguistic information of these rare categories with a pre-trained language model to facilitate learning, thus partially alleviating the problem of insufficient supervision signals. However, the full potential of the textual information and pre-trained language model have been underestimated in the few-shot learning till now, resulting in limited performance enhancements. To address this, we propose a simple but effective framework for few-shot learning tasks, specifically designed to exploit the textual information and language model. In more detail, we explicitly exploit the zero-shot capability of the pre-trained language model with the learnable prompt. And we just add the visual feature with the textual feature for inference directly without the intricate designed fusion modules in previous works. Additionally, we apply the self-ensemble and distillation to further enhance these components. Our extensive experiments conducted across four widely used few-shot datasets demonstrate that our simple framework achieves impressive results. Particularly noteworthy is its outstanding performance in the 1-shot learning task, surpassing state-of-the-art methods by an average of 3.0\% in classification accuracy. \footnote{We will make the source codes of the proposed framework publicly available upon acceptance. }.
♻ ☆ Evaluating Neighbor Explainability for Graph Neural Networks
Explainability in Graph Neural Networks (GNNs) is a new field growing in the last few years. In this publication we address the problem of determining how important is each neighbor for the GNN when classifying a node and how to measure the performance for this specific task. To do this, various known explainability methods are reformulated to get the neighbor importance and four new metrics are presented. Our results show that there is almost no difference between the explanations provided by gradient-based techniques in the GNN domain. In addition, many explainability techniques failed to identify important neighbors when GNNs without self-loops are used.
♻ ☆ Inductive Knowledge Graph Completion with GNNs and Rules: An Analysis
The task of inductive knowledge graph completion requires models to learn inference patterns from a training graph, which can then be used to make predictions on a disjoint test graph. Rule-based methods seem like a natural fit for this task, but in practice they significantly underperform state-of-the-art methods based on Graph Neural Networks (GNNs), such as NBFNet. We hypothesise that the underperformance of rule-based methods is due to two factors: (i) implausible entities are not ranked at all and (ii) only the most informative path is taken into account when determining the confidence in a given link prediction answer. To analyse the impact of these factors, we study a number of variants of a rule-based approach, which are specifically aimed at addressing the aforementioned issues. We find that the resulting models can achieve a performance which is close to that of NBFNet. Crucially, the considered variants only use a small fraction of the evidence that NBFNet relies on, which means that they largely keep the interpretability advantage of rule-based methods. Moreover, we show that a further variant, which does look at the full KG, consistently outperforms NBFNet.
♻ ☆ A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis VLDB 2024
Time series data, including univariate and multivariate ones, are characterized by unique composition and complex multi-scale temporal variations. They often require special consideration of decomposition and multi-scale modeling to analyze. Existing deep learning methods on this best fit to univariate time series only, and have not sufficiently considered sub-series modeling and decomposition completeness. To address these challenges, we propose MSD-Mixer, a Multi-Scale Decomposition MLP-Mixer, which learns to explicitly decompose and represent the input time series in its different layers. To handle the multi-scale temporal patterns and multivariate dependencies, we propose a novel temporal patching approach to model the time series as multi-scale patches, and employ MLPs to capture intra- and inter-patch variations and channel-wise correlations. In addition, we propose a novel loss function to constrain both the mean and the autocorrelation of the decomposition residual for better decomposition completeness. Through extensive experiments on various real-world datasets for five common time series analysis tasks, we demonstrate that MSD-Mixer consistently and significantly outperforms other state-of-the-art algorithms with better efficiency.
comment: Accepted for VLDB 2024
♻ ☆ Point-PEFT: Parameter-Efficient Fine-Tuning for 3D Pre-trained Models
The popularity of pre-trained large models has revolutionized downstream tasks across diverse fields, such as language, vision, and multi-modality. To minimize the adaption cost for downstream tasks, many Parameter-Efficient Fine-Tuning (PEFT) techniques are proposed for language and 2D image pre-trained models. However, the specialized PEFT method for 3D pre-trained models is still under-explored. To this end, we introduce Point-PEFT, a novel framework for adapting point cloud pre-trained models with minimal learnable parameters. Specifically, for a pre-trained 3D model, we freeze most of its parameters, and only tune the newly added PEFT modules on downstream tasks, which consist of a Point-prior Prompt and a Geometry-aware Adapter. The Point-prior Prompt adopts a set of learnable prompt tokens, for which we propose to construct a memory bank with domain-specific knowledge, and utilize a parameter-free attention to enhance the prompt tokens. The Geometry-aware Adapter aims to aggregate point cloud features within spatial neighborhoods to capture fine-grained geometric information through local interactions. Extensive experiments indicate that our Point-PEFT can achieve better performance than the full fine-tuning on various downstream tasks, while using only 5% of the trainable parameters, demonstrating the efficiency and effectiveness of our approach. Code is released at https://github.com/Ivan-Tang-3D/Point-PEFT.
comment: The specialized PEFT framework for 3D pre-trained models, which achieves competitive performance to full fine-tuning, and significantly reduces the computational resources. Project page: https://github.com/Ivan-Tang-3D/Point-PEFT
♻ ☆ Learning Complete Topology-Aware Correlations Between Relations for Inductive Link Prediction
Inductive link prediction -- where entities during training and inference stages can be different -- has shown great potential for completing evolving knowledge graphs in an entity-independent manner. Many popular methods mainly focus on modeling graph-level features, while the edge-level interactions -- especially the semantic correlations between relations -- have been less explored. However, we notice a desirable property of semantic correlations between relations is that they are inherently edge-level and entity-independent. This implies the great potential of the semantic correlations for the entity-independent inductive link prediction task. Inspired by this observation, we propose a novel subgraph-based method, namely TACO, to model Topology-Aware COrrelations between relations that are highly correlated to their topological structures within subgraphs. Specifically, we prove that semantic correlations between any two relations can be categorized into seven topological patterns, and then proposes Relational Correlation Network (RCN) to learn the importance of each pattern. To further exploit the potential of RCN, we propose Complete Common Neighbor induced subgraph that can effectively preserve complete topological patterns within the subgraph. Extensive experiments demonstrate that TACO effectively unifies the graph-level information and edge-level interactions to jointly perform reasoning, leading to a superior performance over existing state-of-the-art methods for the inductive link prediction task.
comment: arXiv admin note: text overlap with arXiv:2103.03642
♻ ☆ From Graph to Word Bag: Introducing Domain Knowledge to Confusing Charge Prediction
Confusing charge prediction is a challenging task in legal AI, which involves predicting confusing charges based on fact descriptions. While existing charge prediction methods have shown impressive performance, they face significant challenges when dealing with confusing charges, such as Snatch and Robbery. In the legal domain, constituent elements play a pivotal role in distinguishing confusing charges. Constituent elements are fundamental behaviors underlying criminal punishment and have subtle distinctions among charges. In this paper, we introduce a novel From Graph to Word Bag (FWGB) approach, which introduces domain knowledge regarding constituent elements to guide the model in making judgments on confusing charges, much like a judge's reasoning process. Specifically, we first construct a legal knowledge graph containing constituent elements to help select keywords for each charge, forming a word bag. Subsequently, to guide the model's attention towards the differentiating information for each charge within the context, we expand the attention mechanism and introduce a new loss function with attention supervision through words in the word bag. We construct the confusing charges dataset from real-world judicial documents. Experiments demonstrate the effectiveness of our method, especially in maintaining exceptional performance in imbalanced label distributions.
♻ ☆ A Literature Review of Literature Reviews in Pattern Analysis and Machine Intelligence
By consolidating scattered knowledge, the literature review provides a comprehensive understanding of the investigated topic. However, reading, conducting, or peer-reviewing review papers generally demands a significant investment of time and effort from researchers. To improve efficiency, this paper aims to provide a thorough review of reviews in the PAMI field from diverse perspectives. First, this paper proposes several article-level, field-normalized, and large language model-empowered bibliometric indicators to evaluate reviews. To facilitate this, a meta-data database dubbed RiPAMI, and a topic dataset are constructed. Second, based on these indicators, the study presents comparative analyses of representative reviews, unveiling the characteristics of publications across various fields, periods, and journals. The newly emerging AI-generated literature reviews are also appraised, and the observed differences suggest that most AI-generated reviews still lag behind human-authored reviews in multiple aspects. Third, we briefly provide a subjective evaluation of representative PAMI reviews and introduce a paper structure-based typology of literature reviews. This typology may improve the clarity and effectiveness for scholars in reading and writing reviews, while also serving as a guide for AI systems in generating well-organized reviews. Finally, this work offers insights into the current challenges of literature reviews and envisions future directions for their development.
comment: IEEE version v1. [February 19, 2024] IEEE version v2 with typos fixed. [February 23, 2024] IEEE version v3 with errors fixed. [February 29, 2024] IEEE version v4 with improved quaility. [February 29, 2024]
♻ ☆ Gradient-less Federated Gradient Boosting Trees with Learnable Learning Rates
The privacy-sensitive nature of decentralized datasets and the robustness of eXtreme Gradient Boosting (XGBoost) on tabular data raise the needs to train XGBoost in the context of federated learning (FL). Existing works on federated XGBoost in the horizontal setting rely on the sharing of gradients, which induce per-node level communication frequency and serious privacy concerns. To alleviate these problems, we develop an innovative framework for horizontal federated XGBoost which does not depend on the sharing of gradients and simultaneously boosts privacy and communication efficiency by making the learning rates of the aggregated tree ensembles learnable. We conduct extensive evaluations on various classification and regression datasets, showing our approach achieves performance comparable to the state-of-the-art method and effectively improves communication efficiency by lowering both communication rounds and communication overhead by factors ranging from 25x to 700x. Project Page: https://flower.ai/blog/2023-04-19-xgboost-with-flower/
comment: Accepted at the 3rd ACM Workshop on Machine Learning and Systems (EuroMLSys), May 8th 2023, Rome, Italy
♻ ☆ UrbanCLIP: Learning Text-enhanced Urban Region Profiling with Contrastive Language-Image Pretraining from the Web
Urban region profiling from web-sourced data is of utmost importance for urban planning and sustainable development. We are witnessing a rising trend of LLMs for various fields, especially dealing with multi-modal data research such as vision-language learning, where the text modality serves as a supplement information for the image. Since textual modality has never been introduced into modality combinations in urban region profiling, we aim to answer two fundamental questions in this paper: i) Can textual modality enhance urban region profiling? ii) and if so, in what ways and with regard to which aspects? To answer the questions, we leverage the power of Large Language Models (LLMs) and introduce the first-ever LLM-enhanced framework that integrates the knowledge of textual modality into urban imagery profiling, named LLM-enhanced Urban Region Profiling with Contrastive Language-Image Pretraining (UrbanCLIP). Specifically, it first generates a detailed textual description for each satellite image by an open-source Image-to-Text LLM. Then, the model is trained on the image-text pairs, seamlessly unifying natural language supervision for urban visual representation learning, jointly with contrastive loss and language modeling loss. Results on predicting three urban indicators in four major Chinese metropolises demonstrate its superior performance, with an average improvement of 6.1% on R^2 compared to the state-of-the-art methods. Our code and the image-language dataset will be released upon paper notification.
comment: Accepted by The Web Conference 2024
♻ ☆ DRL-Based Trajectory Tracking for Motion-Related Modules in Autonomous Driving
Autonomous driving systems are always built on motion-related modules such as the planner and the controller. An accurate and robust trajectory tracking method is indispensable for these motion-related modules as a primitive routine. Current methods often make strong assumptions about the model such as the context and the dynamics, which are not robust enough to deal with the changing scenarios in a real-world system. In this paper, we propose a Deep Reinforcement Learning (DRL)-based trajectory tracking method for the motion-related modules in autonomous driving systems. The representation learning ability of DL and the exploration nature of RL bring strong robustness and improve accuracy. Meanwhile, it enhances versatility by running the trajectory tracking in a model-free and data-driven manner. Through extensive experiments, we demonstrate both the efficiency and effectiveness of our method compared to current methods. Code and documentation are released to facilitate both further research and industrial deployment.
comment: Technical report. Code: https://github.com/MARMOTatZJU/drl-based-trajectory-tracking Documentation: https://drl-based-trajectory-tracking.readthedocs.io
♻ ☆ Enhancing Zero-Shot Chain-of-Thought Reasoning in Large Language Models through Logic COLING 2024
Recent advancements in large language models have showcased their remarkable generalizability across various domains. However, their reasoning abilities still have significant room for improvement, especially when confronted with scenarios requiring multi-step reasoning. Although large language models possess extensive knowledge, their reasoning often fails to effectively utilize this knowledge to establish a coherent thinking paradigm. These models sometimes show hallucinations as their reasoning procedures are unconstrained by logical principles. Aiming at improving the zero-shot chain-of-thought reasoning ability of large language models, we propose LoT (Logical Thoughts), a self-improvement prompting framework that leverages principles rooted in symbolic logic, particularly Reductio ad Absurdum, to systematically verify and rectify the reasoning processes step by step. Experimental evaluations conducted on language tasks in diverse domains, including arithmetic, commonsense, symbolic, causal inference, and social problems, demonstrate the efficacy of enhanced reasoning by logic. The implementation code for LoT can be accessed at: \url{https://github.com/xf-zhao/LoT}.
comment: Accepted in COLING 2024. Code see https://github.com/xf-zhao/LoT
♻ ☆ Self-Supervised Contrastive Learning for Long-term Forecasting ICLR
Long-term forecasting presents unique challenges due to the time and memory complexity of handling long sequences. Existing methods, which rely on sliding windows to process long sequences, struggle to effectively capture long-term variations that are partially caught within the short window (i.e., outer-window variations). In this paper, we introduce a novel approach that overcomes this limitation by employing contrastive learning and enhanced decomposition architecture, specifically designed to focus on long-term variations. To this end, our contrastive loss incorporates global autocorrelation held in the whole time series, which facilitates the construction of positive and negative pairs in a self-supervised manner. When combined with our decomposition networks, our contrastive learning significantly improves long-term forecasting performance. Extensive experiments demonstrate that our approach outperforms 14 baseline models in multiple experiments over nine long-term benchmarks, especially in challenging scenarios that require a significantly long output for forecasting. Source code is available at https://github.com/junwoopark92/Self-Supervised-Contrastive-Forecsating.
comment: Accepted at International Conference on Learning Representations (ICLR) 2024
♻ ☆ A Large-Scale Empirical Study on Improving the Fairness of Image Classification Models
Fairness has been a critical issue that affects the adoption of deep learning models in real practice. To improve model fairness, many existing methods have been proposed and evaluated to be effective in their own contexts. However, there is still no systematic evaluation among them for a comprehensive comparison under the same context, which makes it hard to understand the performance distinction among them, hindering the research progress and practical adoption of them. To fill this gap, this paper endeavours to conduct the first large-scale empirical study to comprehensively compare the performance of existing state-of-the-art fairness improving techniques. Specifically, we target the widely-used application scenario of image classification, and utilized three different datasets and five commonly-used performance metrics to assess in total 13 methods from diverse categories. Our findings reveal substantial variations in the performance of each method across different datasets and sensitive attributes, indicating over-fitting on specific datasets by many existing methods. Furthermore, different fairness evaluation metrics, due to their distinct focuses, yield significantly different assessment results. Overall, we observe that pre-processing methods and in-processing methods outperform post-processing methods, with pre-processing methods exhibiting the best performance. Our empirical study offers comprehensive recommendations for enhancing fairness in deep learning models. We approach the problem from multiple dimensions, aiming to provide a uniform evaluation platform and inspire researchers to explore more effective fairness solutions via a set of implications.
comment: Accepted by the 33rd ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2024). Please include ISSTA in any citations
♻ ☆ A Number Sense as an Emergent Property of the Manipulating Brain
The ability to understand and manipulate numbers and quantities emerges during childhood, but the mechanism through which humans acquire and develop this ability is still poorly understood. We explore this question through a model, assuming that the learner is able to pick up and place small objects from, and to, locations of its choosing, and will spontaneously engage in such undirected manipulation. We further assume that the learner's visual system will monitor the changing arrangements of objects in the scene and will learn to predict the effects of each action by comparing perception with a supervisory signal from the motor system. We model perception using standard deep networks for feature extraction and classification, and gradient descent learning. Our main finding is that, from learning the task of action prediction, an unexpected image representation emerges exhibiting regularities that foreshadow the perception and representation of numbers and quantity. These include distinct categories for zero and the first few natural numbers, a strict ordering of the numbers, and a one-dimensional signal that correlates with numerical quantity. As a result, our model acquires the ability to estimate numerosity, i.e. the number of objects in the scene, as well as subitization, i.e. the ability to recognize at a glance the exact number of objects in small scenes. Remarkably, subitization and numerosity estimation extrapolate to scenes containing many objects, far beyond the three objects used during training. We conclude that important aspects of a facility with numbers and quantities may be learned with supervision from a simple pre-training task. Our observations suggest that cross-modal learning is a powerful learning mechanism that may be harnessed in artificial intelligence.
comment: 16 pages, 5 figures, 15 supplemental figures
♻ ☆ Between Lines of Code: Unraveling the Distinct Patterns of Machine and Human Programmers
Large language models have catalyzed an unprecedented wave in code generation. While achieving significant advances, they blur the distinctions between machine- and human-authored source code, causing integrity and authenticity issues of software artifacts. Previous methods such as DetectGPT have proven effective in discerning machine-generated texts, but they do not identify and harness the unique patterns of machine-generated code. Thus, its applicability falters when applied to code. In this paper, we carefully study the specific patterns that characterize machine- and human-authored code. Through a rigorous analysis of code attributes such as lexical diversity, conciseness, and naturalness, we expose unique patterns inherent to each source. We particularly notice that the syntactic segmentation of code is a critical factor in identifying its provenance. Based on our findings, we propose DetectCodeGPT, a novel method for detecting machine-generated code, which improves DetectGPT by capturing the distinct stylized patterns of code. Diverging from conventional techniques that depend on external LLMs for perturbations, DetectCodeGPT perturbs the code corpus by strategically inserting spaces and newlines, ensuring both efficacy and efficiency. Experiment results show that our approach significantly outperforms state-of-the-art techniques in detecting machine-generated code.
comment: code available at https://github.com/YerbaPage/DetectCodeGPT
Optimization and Control 28
☆ Algorithms of constrained uniform approximation
We address the problem of the best uniform approximation of a continuous function on a convex domain. The approximation is by linear combinations of a finite system of functions (not necessarily Chebyshev) under arbitrary linear constraints. By modifying the concept of alternance and of the Remez iterative procedure we present a method, which demonstrates its efficiency in numerical problems. The linear rate of convergence is proved under some favourable assumptions. A special attention is paid to systems of complex exponents, Gaussian functions, lacunar algebraic and trigonometric polynomials. Applications to signal processing, linear ODE, switching dynamical systems, and to Markov-Bernstein type inequalities are considered.
Optimization on a Finer Scale: Bounded Local Subgradient Variation Perspective
We initiate the study of nonsmooth optimization problems under bounded local subgradient variation, which postulates bounded difference between (sub)gradients in small local regions around points, in either average or maximum sense. The resulting class of objective functions encapsulates the classes of objective functions traditionally studied in optimization, which are defined based on either Lipschitz continuity of the objective or H\"{o}lder/Lipschitz continuity of its gradient. Further, the defined class contains functions that are neither Lipschitz continuous nor have a H\"{o}lder continuous gradient. When restricted to the traditional classes of optimization problems, the parameters defining the studied classes lead to more fine-grained complexity bounds, recovering traditional oracle complexity bounds in the worst case but generally leading to lower oracle complexity for functions that are not ``worst case.'' Some highlights of our results are that: (i) it is possible to obtain complexity results for both convex and nonconvex problems with the (local or global) Lipschitz constant being replaced by a constant of local subgradient variation and (ii) mean width of the subdifferential set around the optima plays a role in the complexity of nonsmooth optimization, particularly in parallel settings. A consequence of (ii) is that for any error parameter $\epsilon > 0$, parallel oracle complexity of nonsmooth Lipschitz convex optimization is lower than its sequential oracle complexity by a factor $\tilde{\Omega}\big(\frac{1}{\epsilon}\big)$ whenever the objective function is piecewise linear with polynomially many pieces in the input size. This is particularly surprising as existing parallel complexity lower bounds are based on such classes of functions. The seeming contradiction is resolved by considering the region in which the algorithm is allowed to query the objective.
☆ Chance-Constrained Gaussian Mixture Steering to a Terminal Gaussian Distribution
We address the problem of finite-horizon control of a discrete-time linear system, where the initial state distribution follows a Gaussian mixture model, the terminal state must follow a specified Gaussian distribution, and the state and control inputs must obey chance constraints. We show that, throughout the time horizon, the state and control distributions are fully characterized by Gaussian mixtures. We then formulate the cost, distributional terminal constraint, and affine/2-norm chance constraints on the state and control, as convex functions of the decision variables. This is leveraged to formulate the chance-constrained path planning problem as a single semidefinite programming problem. A numerical example demonstrates the effectiveness of the proposed method.
☆ Unbiased Extremum Seeking Based on Lie Bracket Averaging IEEE Conference on Decision and Control
Extremum seeking is an online, model-free optimization algorithm traditionally known for its practical stability. This paper introduces an extremum seeking algorithm designed for unbiased convergence to the extremum asymptotically, allowing users to define the convergence rate. Unlike conventional extremum seeking approaches utilizing constant gains, our algorithms employ time-varying parameters. These parameters reduce perturbation amplitudes towards zero in an asymptotic manner, while incorporating asymptotically growing controller gains. The stability analysis is based on state transformation, achieved through the multiplication of the input state by asymptotic growth function, and Lie bracket averaging applied to the transformed system. The averaging ensures the practical stability of the transformed system, which, in turn, leads to the asymptotic stability of the original system. Moreover, for strongly convex maps, we achieve exponentially fast convergence. The numerical simulations validate the feasibility of the introduced designs.
comment: Submitted to the 63rd IEEE Conference on Decision and Control (CDC), 2024
☆ Low Rank Groupwise Deformations for Motion Tracking in Cardiac Cine MRI
Diffeomorphic image registration is a commonly used method to deform one image to resemble another. While warping a single image to another is useful, it can be advantageous to warp multiple images simultaneously, such as in tracking the motion of the heart across a sequence of images. In this paper, our objective is to propose a novel method capable of registering a group or sequence of images to a target image, resulting in registered images that appear identical and therefore have a low rank. Moreover, we aim for these registered images to closely resemble the target image. Through experimental evidence, we will demonstrate our method's superior efficacy in producing low-rank groupwise deformations compared to other state-of-the-art approaches.
comment: A thesis submitted to the University of Birmingham for MSc Degree
☆ Mean Field Game of Mutual Holding with common noise
We consider the mean field game of cross--holding introduced in \citeauthor*{DjeteTouzi} \cite{DjeteTouzi} in the context where the equity value dynamics are affected by a common noise. In contrast with \cite{DjeteTouzi}, the problem exhibits the standard paradigm of mean--variance trade off. Our crucial observation is to search for equilibrium solutions of our mean field game among those models which satisfy an appropriate notion of no--arbitrage. Under this condition, it follows that the representative agent optimization step is reduced to a standard portfolio optimization problem with random endowment.
☆ A Coupled Optimization Framework for Correlated Equilibria in Normal-Form Game
In competitive multi-player interactions, simultaneous optimality is a key requirement for establishing strategic equilibria. This property is explicit when the game-theoretic equilibrium is the simultaneously optimal solution of coupled optimization problems. However, no such optimization problems exist for the correlated equilibrium, a strategic equilibrium where the players can correlate their actions. We address the lack of a coupled optimization framework for the correlated equilibrium by introducing an {unnormalized game} -- an extension of normal-form games in which the player strategies are lifted to unnormalized measures over the joint actions. We show that the set of fully mixed generalized Nash equilibria of this unnormalized game is a subset of the correlated equilibrium of the normal-form game. Furthermore, we introduce an entropy regularization to the unnormalized game and prove that the entropy-regularized generalized Nash equilibrium is a sub-optimal correlated equilibrium of the normal form game where the degree of sub-optimality depends on the magnitude of regularization. We prove that the entropy-regularized unnormalized game has a closed-form solution, and empirically verify its computational efficacy at approximating the correlated equilibrium of normal-form games.
comment: 8 pages, 2 figures
☆ A robust optimization approach model for a multi-vaccine multi-echelon supply chain
This research investigates a multi-product, multi-echelon, and multi-period vaccine supply chain network model under uncertainty and quality inspection errors. The objective function seeks optimizing the total cost of the supply chain. Moreover, the proposed model is formulated as a mixed integer linear programming problem under multiple sources of uncertain parameters including demand, inspection errors, vaccine waste generated in healthcare centers, and defective treatment rate of vaccine waste. To provide meaningful solutions that are robust against future fluctuation of parameters, the robust optimization approach is utilized to incorporate the decision maker risk attitude under different type of uncertainty sets. Namely, box, polyhedral and combination of interval polyhedral. The performance of the proposed model is demonstrated through an illustrative example. The results show the effect of different types of uncertainties on the overall objective function. Managerial insights and research implications in terms of vaccine supply chain is advised and future research directions are proposed.
☆ Input-to-State Stability of Newton Methods for Generalized Equations in Nonlinear Optimization
We show that Newton methods for generalized equations are input-to-state stable with respect to disturbances such as due to inexact computations. We then use this result to obtain convergence and robustness of a multistep Newton-type method for multivariate generalized equations. We demonstrate the usefulness of the results with other applications to nonlinear optimization. In particular, we provide a new proof for (robust) local convergence of the augmented Lagrangian method.
comment: Submitted to 2024 Conference on Decision and Control
☆ An instability result of Hamiltonian systems related to optimal swing-up control of a pendulum IEEE Conference on Decision and Control 2024
This paper presents an instability result of Hamiltonian systems associated with optimal swing-up control for a pendulum. The systems possess weak (higher-order) instability at the initial point of the swing-up control, the analysis for which requires techniques from celestial mechanics. The obtained result may have relationships with the previously obtained numerical studies for the existence of multiple locally optimal solutions and the non-existence conjecture of the optimal control.
comment: Submitted for the IEEE Conference on Decision and Control 2024
☆ On the Bailout Dividend Problem with Periodic Dividend Payments and Fixed Transaction Costs
We study the optimal bailout dividend problem with transaction costs for an insurance company, where shareholder payouts align with the arrival times of an independent Poisson process. In this scenario, the underlying risk model follows a spectrally negative L\'evy process. Our analysis confirms the optimality of a periodic $(b_{1},b_{2})$-barrier policy with classical reflection at zero. This strategy involves reducing the surplus to $b_1$ when it exceeds $b_{2}$ at the Poisson arrival times and pushes the surplus to 0 whenever it goes below zero.
☆ Towards a MATLAB Toolbox to compute backstepping kernels using the power series method
In this paper, we extend our previous work on the power series method for computing backstepping kernels. Our first contribution is the development of initial steps towards a MATLAB toolbox dedicated to backstepping kernel computation. This toolbox would exploit MATLAB's linear algebra and sparse matrix manipulation features for enhanced efficiency; our initial findings show considerable improvements in computational speed with respect to the use of symbolical software without loss of precision at high orders. Additionally, we tackle limitations observed in our earlier work, such as slow convergence (due to oscillatory behaviors) and non-converging series (due to loss of analiticity at some singular points). To overcome these challenges, we introduce a technique that mitigates this behaviour by computing the expansion at different points, denoted as localized power series. This approach effectively navigates around singularities, and can also accelerates convergence by using more local approximations. Basic examples are provided to demonstrate these enhancements. Although this research is still ongoing, the significant potential and simplicity of the method already establish the power series approach as a viable and versatile solution for solving backstepping kernel equations, benefiting both novel and experienced practitioners in the field. We anticipate that these developments will be particularly beneficial in training the recently introduced neural operators that approximate backstepping kernels and gains.
comment: Preprint submitted to CDC 2024
☆ Manifold Regularization Classification Model Based On Improved Diffusion Map
Manifold regularization model is a semi-supervised learning model that leverages the geometric structure of a dataset, comprising a small number of labeled samples and a large number of unlabeled samples, to generate classifiers. However, the original manifold norm limits the performance of models to local regions. To address this limitation, this paper proposes an approach to improve manifold regularization based on a label propagation model. We initially enhance the probability transition matrix of the diffusion map algorithm, which can be used to estimate the Neumann heat kernel, enabling it to accurately depict the label propagation process on the manifold. Using this matrix, we establish a label propagation function on the dataset to describe the distribution of labels at different time steps. Subsequently, we extend the label propagation function to the entire data manifold. We prove that the extended label propagation function converges to a stable distribution after a sufficiently long time and can be considered as a classifier. Building upon this concept, we propose a viable improvement to the manifold regularization model and validate its superiority through experiments.
comment: 20 pages, 24figures
☆ Exponential mixing of constrained random dynamical systems via controllability conditions
We provide deterministic controllability conditions that imply exponential mixing properties for randomly forced constrained dynamical systems with possibly unbounded state space. As an application, new ergodicity results are obtained for non-smooth models in elasto-plasticity driven by various types of noise, including white noise. It is thereby illustrated how tools from control theory can be utilized to tackle regularity issues that commonly arise in the qualitative study of constrained systems.
comment: 27 pages, 1 figure
☆ Digital control of negative imaginary systems: a discrete-time hybrid integrator-gain system approach
A hybrid integrator-gain system (HIGS) is a control element that switches between an integrator and a gain, which overcomes some inherent limitations of linear controllers. In this paper, we consider using discrete-time HIGS controllers for the digital control of negative imaginary (NI) systems. We show that the discrete-time HIGS themselves are step-advanced negative imaginary systems. For a minimal linear NI system, there always exists a HIGS controller that can asymptotically stablize it. An illustrative example is provided, where we use the proposed HIGS control method to stabilize a discrete-time mass-spring system.
comment: To appear in the 2024 European Control Conference. 7 pages, 3 figures
☆ Planning Charging Stations and Service Operations of Dockless Electric Micromobility Systems
Dockless electric micro-mobility services (e.g., shared e-scooters and e-bikes) have been increasingly popular in the recent decade, and a variety of charging technologies have emerged for these services. The use of charging stations, to/from which service vehicles are transported by the riders for charging, poses as a promising approach because it reduces the need for dedicated staff or contractors. However, unique challenges also arise, such as how to incentivize riders to drop off vehicles at stations and how to efficiently utilize the vehicles being charged at the stations. This paper focuses on dockless e-scooters as an example and develops a new spatial queuing network model to capture the steady-state scooter service cycles, battery consumption and charging processes, and the associated pricing and management mechanisms. Building upon this model, a system of closed-form equations is formulated and incorporated into a constrained nonlinear program to optimize the deployment of the service fleet, the design of charging stations (i.e., number, location, and capacity), user-based charging price promotions and priorities, and repositioning truck operations (i.e., headway and truck load). The proposed queuing network model is found to match very well with agent-based simulations. It is applied to a series of numerical experiments to draw insights into the optimal designs and the system performance. The numerical results reveal strong advantages of using charging stations for shared dockless electric micro-mobility services as compared to state-of-the-art alternatives. The proposed model can also be used to analyze other micromobility services and other charging approaches.
☆ Stochastic maximum principle for weighted mean-field system with jump
In this article, we consider a weighted mean-field control problem with jump-diffusion as its state process. The main difficulty is from the non-Lipschitz property of the coefficients. We overcome this difficulty by an $L_{p,q}$-estimate of the solution processes with a suitably chosen $p$ and $q$. Convex pertubation method combining with the aforementioned $L_{p,q}$-estimation method is utilized to derive the stochastic maximum principle for this control problem. A sufficient condition for the optimality is also given.
☆ Output Feedback Control of Suspended Sediment Load Entrainment in Water Canals and Reservoirs
This paper addresses the management of water flow in a rectangular open channel, considering the dynamic nature of both the channel's bathymetry and the suspended sediment particles caused by entrainment and deposition effects. The control-oriented model under study is a set of coupled nonlinear partial differential equations (PDEs) describing conservation of mass and momentum while accounting for constitutive relations that govern sediment erosion and deposition phenomena. The proposed boundary control problem presents a fresh perspective in water canal management and expands Saint-Venant Exner (SVE) control frameworks by integrating dynamics related to the transport of fine particles. After linearization, PDE backstepping design is employed to stabilize both the bathymetry, the water dynamics together with the concentration of suspended sediment particles. Two underflow sluice gates are used for flow control at the upstream and downstream boundaries with only the downstream component being actuated. An observer-based backstepping control design is carried out for the downstream gate using state measurement at the upstream gate to globally exponentially stabilize the linearized system to a desired equilibrium point in $\mathscr{L}^2$ sense. The stability analysis is performed on the linearized model which is a system of four coupled PDEs, three of which are rightward convecting and one leftward. The proposed control design has the potential to facilitate efficient reservoir flushing operations. Consistent simulation results are presented to illustrate the feasibility of the designed control law.
comment: 10 pages, 8 figures
☆ Infinite dimensional open-loop linear quadratic stochastic optimal control problems and related games
We investigate the linear quadratic stochastic optimal control problems in infinite dimension without Markovian restriction for coefficients. The necessary and sufficient conditions for open-loop optimal controls are presented. We prove the Fr\'echet differentiable of the cost functional with respect to the control variable, and the Fr\'echet derivatives are characterized in detail by operators derived from dual analysis, which are proven to be the stationary conditions. Transposition methods are adopted to deal with the adjoint equations. As applications, we employ the results to study open-loop Nash equilibria for two-person stochastic differential games.
comment: 16 pages
♻ ☆ Stochastic Hessian Fittings on Lie Groups
This paper studies the fitting of Hessian or its inverse for stochastic optimizations using a Hessian fitting criterion from the preconditioned stochastic gradient descent (PSGD) method, which is intimately related to many commonly used second order and adaptive gradient optimizers, e.g., BFGS, Gaussian-Newton and natural gradient descent, AdaGrad, etc. Our analyses reveal the efficiency and reliability differences among a wide range of preconditioner fitting methods, from closed-form to iterative solutions, using Hessian-vector products or stochastic gradients only, with Hessian fittings in the Euclidean space, the manifold of symmetric positive definite (SPL) matrices, or a variety of Lie groups. The most intriguing discovery is that the Hessian fitting itself as an optimization problem is strongly convex under mild conditions on a specific yet general enough Lie group. This discovery turns Hessian fitting into a well behaved optimization problem, and facilitates the designs of highly efficient and elegant Lie group sparse preconditioner fitting methods for large scale stochastic optimizations.
♻ ☆ The chain control set of discrete-time linear system on the affine two-dimensional Lie group
In this paper, we present conditions for the existence and uniqueness of chain control sets of discrete-time linear systems on the affine two-dimensional Lie group. More specifically, we prove that these chain control sets are given by the union of an infinite number of control sets with empty interiors.
♻ ☆ A single player and a mass of agents: a pursuit evasion-like game
We study a finite-horizon differential game of pursuit-evasion like, between a single player and a mass of agents. The player and the mass directly control their own evolution, which for the mass is given by a first order PDE of transport equation type. Using also an adapted concept of non-anticipating strategies, we derive an infinite dimensional Isaacs equation, and by dynamic programming techniques we prove that the value function is the unique viscosity solution on a suitable invariant subset of a Hilbert space.
♻ ☆ Adaptive Softassign via Hadamard-Equipped Sinkhorn
Softassign is a pivotal method in graph matching and other learning tasks. Many softassign-based algorithms exhibit performance sensitivity to a parameter in the softassign. However, tuning the parameter is challenging and almost done empirically. This paper proposes an adaptive softassign method for graph matching by analyzing the relationship between the objective score and the parameter. This method can automatically tune the parameter based on a given error bound to guarantee accuracy. The Hadamard-Equipped Sinkhorn formulas introduced in this study significantly enhance the efficiency and stability of the adaptive softassign. Moreover, these formulas can also be used in optimal transport problems. The resulting adaptive softassign graph matching algorithm enjoys significantly higher accuracy than previous state-of-the-art large graph matching algorithms while maintaining comparable efficiency.
♻ ☆ Exact controllability of incompressible ideal magnetohydrodynamics in $2$D
This work examines the controllability of planar incompressible ideal magnetohydrodynamics (MHD). Interior controls are obtained for problems posed in doubly-connected regions; simply-connected configurations are driven by boundary controls. Up to now, only straight channels regulated at opposing walls have been studied. Hence, the present program adds to the literature an exploration of interior controllability, extends the known boundary controllability results, and contributes ideas for treating general domains. To transship obstacles stemming from the MHD coupling and the magnetic field topology, a divide-and-control strategy is proposed. This leads to a family of nonlinear velocity-controlled sub-problems which are solved using J.-M. Coron's return method. The latter is here developed based on a reference trajectory in the domain's first cohomology space.
comment: 43 pages, 9 figures
♻ ☆ Exponential Concentration in Stochastic Approximation
We analyze the behavior of stochastic approximation algorithms where iterates, in expectation, progress towards an objective at each step. When progress is proportional to the step size of the algorithm, we prove exponential concentration bounds. These tail-bounds contrast asymptotic normality results, which are more frequently associated with stochastic approximation. The methods that we develop rely on a geometric ergodicity proof. This extends a result on Markov chains due to Hajek (1982) to the area of stochastic approximation algorithms. We apply our results to several different Stochastic Approximation algorithms, specifically Projected Stochastic Gradient Descent, Kiefer-Wolfowitz and Stochastic Frank-Wolfe algorithms. When applicable, our results prove faster $O(1/t)$ and linear convergence rates for Projected Stochastic Gradient Descent with a non-vanishing gradient.
comment: 35 pages, 11 Figures
♻ ☆ An adjoint-free algorithm for conditional nonlinear optimal perturbations (CNOPs) via sampling
In this paper, we propose a sampling algorithm based on state-of-the-art statistical machine learning techniques to obtain conditional nonlinear optimal perturbations (CNOPs), which is different from traditional (deterministic) optimization methods.1 Specifically, the traditional approach is unavailable in practice, which requires numerically computing the gradient (first-order information) such that the computation cost is expensive, since it needs a large number of times to run numerical models. However, the sampling approach directly reduces the gradient to the objective function value (zeroth-order information), which also avoids using the adjoint technique that is unusable for many atmosphere and ocean models and requires large amounts of storage. We show an intuitive analysis for the sampling algorithm from the law of large numbers and further present a Chernoff-type concentration inequality to rigorously characterize the degree to which the sample average probabilistically approximates the exact gradient. The experiments are implemented to obtain the CNOPs for two numerical models, the Burgers equation with small viscosity and the Lorenz-96 model. We demonstrate the CNOPs obtained with their spatial patterns, objective values, computation times, and nonlinear error growth. Compared with the performance of the three approaches, all the characters for quantifying the CNOPs are nearly consistent, while the computation time using the sampling approach with fewer samples is much shorter. In other words, the new sampling algorithm shortens the computation time to the utmost at the cost of losing little accuracy.
comment: 20 pages, 6 figures, 4 tables
♻ ☆ Learning Acceptability Criteria from Good and Bad Decisions: An Inverse Optimization Approach
Conventional inverse optimization inputs a solution and finds the parameters of an optimization model that render a given solution optimal. The literature mostly focuses on inferring the objective function in linear problems when acceptable solutions are provided as input. In this paper, we propose an inverse optimization model that inputs several accepted and rejected solutions and recovers the underlying convex optimization model that can be used to generate such solutions. The novelty of our model is three-fold: First, while most literature focuses on inferring the objective function, we focus on inferring the feasible region. Second, our model can infer the constraints of general convex optimization models. Third, the proposed model learns from accepted (good) and rejected (bad) observations in inferring the constraint set. The resulting inverse model is a mixed-integer nonlinear problem that is complex to solve. To mitigate the inverse problem complexity, we employ variational inequalities and the theoretical properties of the solutions to derive a reduced formulation that retains the complexity of its forward counterpart. We demonstrate that our inverse model can utilize a subset of past good and bad treatment plans to infer planning criteria that can lead to nearly guaranteed clinically acceptable plans for future patients.
♻ ☆ Best of Both Worlds Guarantees for Smoothed Online Quadratic Optimization
We study the smoothed online quadratic optimization (SOQO) problem where, at each round $t$, a player plays an action $x_t$ in response to a quadratic hitting cost and an additional squared $\ell_2$-norm cost for switching actions. This problem class has strong connections to a wide range of application domains including smart grid management, adaptive control, and data center management, where switching-efficient algorithms are highly sought after. We study the SOQO problem in both adversarial and stochastic settings, and in this process, perform the first stochastic analysis of this class of problems. We provide the online optimal algorithm when the minimizers of the hitting cost function evolve as a general stochastic process, which, for the case of martingale process, takes the form of a distribution-agnostic dynamic interpolation algorithm (LAI). Next, we present the stochastic-adversarial trade-off by proving an $\Omega(T)$ expected regret for the adversarial optimal algorithm in the literature (ROBD) with respect to LAI and, a sub-optimal competitive ratio for LAI in the adversarial setting. Finally, we present a best-of-both-worlds algorithm that obtains a robust adversarial performance while simultaneously achieving a near-optimal stochastic performance.
comment: 48 pages, 9 figures
Systems and Control 21
Datasets of Great Britain Primary Substations Integrated with Household Heating Information
The growing demand for electrified heating, electrified transportation, and power-intensive data centres challenge distribution networks. If electrification projects are carried out without considering electrical distribution infrastructure, there could be unexpected blackouts and financial losses. Datasets containing real-world distribution network information are required to address this. On the other hand, social data, such as household heating composition, are closely coupled with people's lives. Studying the coupling between the energy system and society is important in promoting social welfare. To fill these gaps, this paper introduces two datasets. The first is the main dataset for the distribution networks in Great Britain (GB), collecting information on firm capacity, peak demands, locations, and parent transmission nodes (the Grid Supply Point, namely GSP) for all primary substations (PSs). PSs are a crucial part of the UK distribution network and are at the lowest voltage level (11 kV) with publicly available data for most UK Distribution Network Operators (DNOs). Substation firm capacity and peak demand facilitate an understanding of the remaining room of the existing network. The parent GSP information helps link the dataset of distribution networks to datasets of transmission networks. The second dataset extends the main network dataset, linking each PS to information about the number of households that use different types of central heating recorded in census data. The derivation of the second dataset is based on locations of PSs collected in the main dataset with appropriate assumptions. The derivation process may also be replicated to integrate other social datasets.
comment: Submitted to the journal "Data in Brief"
☆ ANN-Based Adaptive NMPC for Uranium Extraction-Scrubbing Operation in Spent Nuclear Fuel Treatment Process
This paper addresses the particularities in optimal control of the uranium extraction-scrubbing operation in the PUREX process. The control problem requires optimally stabilizing the system at a desired solvent saturation level, guaranteeing constraints, disturbance rejection, and adapting to set point variations. A qualified simulator named PAREX was developed by the French Alternative Energies and Atomic Energy Commission (CEA) to simulate liquid-liquid extraction operations in the PUREX process. However, since the mathematical model is complex and is described by a system of nonlinear, stiff, high-dimensional differential-algebraic equations (DAE), applying optimal control methods will lead to a large-scale nonlinear programming problem with a huge computational burden. The solution we propose in this work is to train a neural network to predict the process outputs using the measurement history. This neural network architecture, which employs the long short-term memory (LSTM), linear regression and logistic regression networks, allows reducing the number of state variables, thus reducing the complexity of the optimization problems in the control scheme. Furthermore, nonlinear model predictive control (NMPC) and moving horizon estimation (MHE) problems are developed and solved using the PSO (Particle Swarm Optimization) algorithm. Simulation results show that the proposed adaptive optimal control scheme satisfies the requirements of the control problem and provides promise for experimental testing.
Control-Coherent Koopman Modeling: A Physical Modeling Approach
The modeling of nonlinear dynamics based on Koopman operator theory, which is originally applicable only to autonomous systems with no control, is extended to non-autonomous control system without approximation to input matrix B. Prevailing methods using a least square estimate of the B matrix may result in an erroneous input matrix, misinforming the controller about the structure of the input matrix in a lifted space. Here, a new method for constructing a Koopman model that comprises the exact input matrix B is presented. A set of state variables are introduced so that the control inputs are linearly involved in the dynamics of actuators. With these variables, a lifted linear model with the exact control matrix, called a Control-Coherent Koopman Model, is constructed by superposing control input terms, which are linear in local actuator dynamics, to the Koopman operator of the associated autonomous nonlinear system. The proposed method is applied to multi degree-of-freedom robotic arms and multi-cable manipulation systems. Model Predictive Control is applied to the former. It is demonstrated that the prevailing Dynamic Mode Decomposition with Control (DMDc) using an approximate control matrix B does not provide a satisfactory result, while the Control-Coherent Koopman Model performs well with the correct B matrix.
☆ Semi-Automatic Line-System Provisioning with Integrated Physical-Parameter-Aware Methodology: Field Verification and Operational Feasibility
We propose methods and an architecture to conduct measurements and optimize newly installed optical fiber line systems semi-automatically using integrated physics-aware technologies in a data center interconnection (DCI) transmission scenario. We demonstrate, for the first time, digital longitudinal monitoring (DLM) and optical line system (OLS) physical parameter calibration working together in real-time to extract physical link parameters for transmission performance optimization. Our methodology has the following advantages over traditional design: a minimized footprint at user sites, accurate estimation of the necessary optical network characteristics via complementary telemetry technologies, and the capability to conduct all operation work remotely. The last feature is crucial, as it enables remote operation to implement network design settings for immediate response to quality of transmission (QoT) degradation and reversion in the case of unforeseen problems. We successfully performed semi-automatic line system provisioning over field fiber networks facilities at Duke University, Durham, NC. The tasks of parameter retrieval, equipment setting optimization, and system setup/provisioning were completed within 1 hour. The field operation was supervised by on-duty personnel who could access the system remotely from different time zones. By comparing Q-factor estimates calculated from the extracted link parameters with measured results from 400G transceivers, we confirmed that our methodology has a reduction in the QoT prediction errors (+-0.3 dB) over existing design (+-10.6 dB).
☆ HT-LIP Model based Robust Control of Quadrupedal Robot Locomotion under Unknown Vertical Ground Motion
This paper presents a hierarchical control framework that enables robust quadrupedal locomotion on a dynamic rigid surface (DRS) with general and unknown vertical motions. The key novelty of the framework lies in its higher layer, which is a discrete-time, provably stabilizing footstep controller. The basis of the footstep controller is a new hybrid, time-varying, linear inverted pendulum (HT-LIP) model that is low-dimensional and accurately captures the essential robot dynamics during DRS locomotion. A new set of sufficient stability conditions are then derived to directly guide the controller design for ensuring the asymptotic stability of the HT-LIP model under general, unknown, vertical DRS motions. Further, the footstep controller is cast as a computationally efficient quadratic program that incorporates the proposed HT-LIP model and stability conditions. The middle layer takes the desired footstep locations generated by the higher layer as input to produce kinematically feasible full-body reference trajectories, which are then accurately tracked by a lower-layer torque controller. Hardware experiments on a Unitree Go1 quadrupedal robot confirm the robustness of the proposed framework under various unknown, aperiodic, vertical DRS motions and uncertainties (e.g., slippery and uneven surfaces, solid and liquid loads, and sudden pushes).
Legged Robot State Estimation within Non-inertial Environments
This paper investigates the robot state estimation problem within a non-inertial environment. The proposed state estimation approach relaxes the common assumption of static ground in the system modeling. The process and measurement models explicitly treat the movement of the non-inertial environments without requiring knowledge of its motion in the inertial frame or relying on GPS or sensing environmental landmarks. Further, the proposed state estimator is formulated as an invariant extended Kalman filter (InEKF) with the deterministic part of its process model obeying the group-affine property, leading to log-linear error dynamics. The observability analysis of the filter confirms that the robot's pose (i.e., position and orientation) and velocity relative to the non-inertial environment are observable. Hardware experiments on a humanoid robot moving on a rotating and translating treadmill demonstrate the high convergence rate and accuracy of the proposed InEKF even under significant treadmill pitch sway, as well as large estimation errors.
☆ Bi-Level Control of Weaving Sections in Mixed Traffic Environments with Connected and Automated Vehicles
Connected and automated vehicles (CAVs) can be beneficial for improving the operation of highway bottlenecks such as weaving sections. This paper proposes a bi-level control approach based on an upper-level deep reinforcement learning controller and a lower-level model predictive controller to coordinate the lane-changings of a mixed fleet of CAVs and human-driven vehicles (HVs) in weaving sections. The upper level represents a roadside controller that collects vehicular information from the entire weaving section and determines the control weights used in the lower-level controller. The lower level is implemented within each CAV, which takes the control weights from the upper-level controller and generates the acceleration and steering angle for individual CAVs based on the local situation. The lower-level controller further incorporates an HV trajectory predictor, which is capable of handling the dynamic topology of vehicles in weaving scenarios with intensive mandatory lane changes. The case study inspired by a real weaving section in Basel, Switzerland, shows that our method consistently outperforms state-of-the-art benchmarks.
comment: 12 pages, 8 figures
☆ Efficient Reachable Sets on Lie Groups Using Lie Algebra Monotonicity and Tangent Intervals
In this paper, we efficiently compute overapproximated reachable sets for control systems evolving on Lie groups, building off results from monotone systems theory and geometric integration theory. We propose to consider intervals living in the Lie algebra, which through the exponential map, describe real sets on the Lie group. A local equivalence between the original system and a system evolving on the Lie algebra allows existing interval reachability techniques to apply in the tangent space. Using interval bounds of the Baker-Campbell-Hausdorff formula, a Runge-Kutta-Munthe-Kaas reachability algorithm is proposed, providing reachable set estimates for arbitrary time horizons at little computational cost. The algorithm is demonstrated on through consensus on a torus and attitude control on $SO(3)$.
☆ Voltage Regulation in Polymer Electrolyte Fuel Cell Systems Using Gaussian Process Model Predictive Control
This study introduces a novel approach utilizing Gaussian process model predictive control (MPC) to stabilize the output voltage of a polymer electrolyte fuel cell (PEFC) system by simultaneously regulating hydrogen and airflow rates. Two Gaussian process models are developed to capture PEFC dynamics, taking into account constraints including hydrogen pressure and input change rates, thereby aiding in mitigating errors inherent to PEFC predictive control. The dynamic performance of the physical model and Gaussian process MPC in constraint handling and system inputs is compared and analyzed. Simulation outcomes demonstrate that the proposed Gaussian process MPC effectively maintains the voltage at the target 48 V while adhering to safety constraints, even amidst workload disturbances ranging from 110-120 A. In comparison to traditional MPC using detailed system models, Gaussian process MPC exhibits a 43\% higher overshoot and 25\% slower response time. Nonetheless, it offers the advantage of not requiring the underlying true system model and needing less system information.
☆ Input-to-State Stability of Newton Methods for Generalized Equations in Nonlinear Optimization
We show that Newton methods for generalized equations are input-to-state stable with respect to disturbances such as due to inexact computations. We then use this result to obtain convergence and robustness of a multistep Newton-type method for multivariate generalized equations. We demonstrate the usefulness of the results with other applications to nonlinear optimization. In particular, we provide a new proof for (robust) local convergence of the augmented Lagrangian method.
comment: Submitted to 2024 Conference on Decision and Control
☆ Data-Driven Sliding Mode Control for Partially Unknown Nonlinear Systems
This paper introduces a new design method for data-driven control of nonlinear systems with partially unknown dynamics and unknown bounded disturbance. Since it is not possible to achieve exact nonlinearity cancellation in the presence of unknown disturbance, this paper adapts the idea of sliding mode control (SMC) to ensure system stability and robustness without assuming that the nonlinearity goes to zero faster than the state as in the existing methods. The SMC consists of a data-dependent robust controller ensuring the system state trajectory reach and remain on the sliding surface and a nominal controller solved from a data-dependent semidefinite program (SDP) ensuring robust stability of the state trajectory on the sliding surface. Numerical simulation results demonstrate effectiveness of the proposed data-driven SMC and its superior in terms of robust stability over the existing data-driven control that also uses approximate nonlinearity cancellation.
comment: Submitted to IEEE CDC 2024
☆ Runtime Monitoring and Fault Detection for Neural Network-Controlled Systems
There is an emerging trend in applying deep learning methods to control complex nonlinear systems. This paper considers enhancing the runtime safety of nonlinear systems controlled by neural networks in the presence of disturbance and measurement noise. A robustly stable interval observer is designed to generate sound and precise lower and upper bounds for the neural network, nonlinear function, and system state. The obtained interval is utilised to monitor the real-time system safety and detect faults in the system outputs or actuators. An adaptive cruise control vehicular system is simulated to demonstrate effectiveness of the proposed design.
comment: Accepted to SAFEPROCESS 2024
☆ Towards a MATLAB Toolbox to compute backstepping kernels using the power series method
In this paper, we extend our previous work on the power series method for computing backstepping kernels. Our first contribution is the development of initial steps towards a MATLAB toolbox dedicated to backstepping kernel computation. This toolbox would exploit MATLAB's linear algebra and sparse matrix manipulation features for enhanced efficiency; our initial findings show considerable improvements in computational speed with respect to the use of symbolical software without loss of precision at high orders. Additionally, we tackle limitations observed in our earlier work, such as slow convergence (due to oscillatory behaviors) and non-converging series (due to loss of analiticity at some singular points). To overcome these challenges, we introduce a technique that mitigates this behaviour by computing the expansion at different points, denoted as localized power series. This approach effectively navigates around singularities, and can also accelerates convergence by using more local approximations. Basic examples are provided to demonstrate these enhancements. Although this research is still ongoing, the significant potential and simplicity of the method already establish the power series approach as a viable and versatile solution for solving backstepping kernel equations, benefiting both novel and experienced practitioners in the field. We anticipate that these developments will be particularly beneficial in training the recently introduced neural operators that approximate backstepping kernels and gains.
comment: Preprint submitted to CDC 2024
☆ Digital control of negative imaginary systems: a discrete-time hybrid integrator-gain system approach
A hybrid integrator-gain system (HIGS) is a control element that switches between an integrator and a gain, which overcomes some inherent limitations of linear controllers. In this paper, we consider using discrete-time HIGS controllers for the digital control of negative imaginary (NI) systems. We show that the discrete-time HIGS themselves are step-advanced negative imaginary systems. For a minimal linear NI system, there always exists a HIGS controller that can asymptotically stablize it. An illustrative example is provided, where we use the proposed HIGS control method to stabilize a discrete-time mass-spring system.
comment: To appear in the 2024 European Control Conference. 7 pages, 3 figures
☆ Force Controlled Printing for Material Extrusion Additive Manufacturing
In material extrusion additive manufacturing, the extrusion process is commonly controlled in a feed-forward fashion. The amount of material to be extruded at each printing location is pre-computed by a planning software. This approach is inherently unable to adapt the extrusion to external and unexpected disturbances, and the quality of the results strongly depends on a number of modeling and tuning parameters. To overcome these limitations, we propose the first framework for Force Controlled Printing for material extrusion additive manufacturing. We utilize a custom-built extruder to measure the extrusion force in real time, and use this quantity as feedback to continuously control the material flow in closed-loop. We demonstrate the existence of a strong correlation between extrusion force and line width, which we exploit to deposit lines of desired width in a width range of 33 % up to 233 % of the nozzle diameter. We also show how Force Controlled Printing outperforms conventional feed-forward extrusion in print quality and disturbance rejection, while requiring little tuning and automatically adapting to changes in the hardware settings. With no adaptation, Force Controlled Printing can deposit lines of desired width under severe disturbances in bed leveling, such as at layer heights ranging between 20 % and 200 % of the nominal height.
comment: Preprint to be submitted to Elsevier Additive Manufacturing
☆ Fisher Information Approach for Masking the Sensing Plan: Applications in Multifunction Radars
How to design a Markov Decision Process (MDP) based radar controller that makes small sacrifices in performance to mask its sensing plan from an adversary? The radar controller purposefully minimizes the Fisher information of its emissions so that an adversary cannot identify the controller's model parameters accurately. Unlike classical open loop statistical inference, where the Fisher information serves as a lower bound for the achievable covariance, this paper employs the Fisher information as a design constraint for a closed loop radar controller to mask its sensing plan. We analytically derive a closed-form expression for the determinant of the Fisher Information Matrix (FIM) pertaining to the parameters of the MDP-based controller. Subsequently, we constrain the MDP with respect to the determinant of the FIM. Numerical results show that the introduction of minor perturbations to the MDP's transition kernel and the total operation cost can reduce the Fisher Information of the emissions. Consequently, this reduction amplifies the variability in policy and transition kernel estimation errors, thwarting the adversary's accuracy in estimating the controller's sensing plan.
♻ ☆ Concurrent Learning of Policy and Unknown Safety Constraints in Reinforcement Learning
Reinforcement learning (RL) has revolutionized decision-making across a wide range of domains over the past few decades. Yet, deploying RL policies in real-world scenarios presents the crucial challenge of ensuring safety. Traditional safe RL approaches have predominantly focused on incorporating predefined safety constraints into the policy learning process. However, this reliance on predefined safety constraints poses limitations in dynamic and unpredictable real-world settings where such constraints may not be available or sufficiently adaptable. Bridging this gap, we propose a novel approach that concurrently learns a safe RL control policy and identifies the unknown safety constraint parameters of a given environment. Initializing with a parametric signal temporal logic (pSTL) safety specification and a small initial labeled dataset, we frame the problem as a bilevel optimization task, intricately integrating constrained policy optimization, using a Lagrangian-variant of the twin delayed deep deterministic policy gradient (TD3) algorithm, with Bayesian optimization for optimizing parameters for the given pSTL safety specification. Through experimentation in comprehensive case studies, we validate the efficacy of this approach across varying forms of environmental constraints, consistently yielding safe RL policies with high returns. Furthermore, our findings indicate successful learning of STL safety constraint parameters, exhibiting a high degree of conformity with true environmental safety constraints. The performance of our model closely mirrors that of an ideal scenario that possesses complete prior knowledge of safety constraints, demonstrating its proficiency in accurately identifying environmental safety constraints and learning safe policies that adhere to those constraints.
♻ ☆ Efficient Recursive Data-enabled Predictive Control (Extended Version)
In the field of model predictive control, Data-enabled Predictive Control (DeePC) offers direct predictive control, bypassing traditional modeling. However, challenges emerge with increased computational demand due to recursive data updates. This paper introduces a novel recursive updating algorithm for DeePC. It emphasizes the use of Singular Value Decomposition (SVD) for efficient low-dimensional transformations of DeePC in its general form, as well as a fast SVD update scheme. Importantly, our proposed algorithm is highly flexible due to its reliance on the general form of DeePC, which is demonstrated to encompass various data-driven methods that utilize Pseudoinverse and Hankel matrices. This is exemplified through a comparison to Subspace Predictive Control, where the algorithm achieves asymptotically consistent prediction for stochastic linear time-invariant systems. Our proposed methodologies' efficacy is validated through simulation studies.
♻ ☆ Hierarchical Federated Learning in Wireless Networks: Pruning Tackles Bandwidth Scarcity and System Heterogeneity
While a practical wireless network has many tiers where end users do not directly communicate with the central server, the users' devices have limited computation and battery powers, and the serving base station (BS) has a fixed bandwidth. Owing to these practical constraints and system models, this paper leverages model pruning and proposes a pruning-enabled hierarchical federated learning (PHFL) in heterogeneous networks (HetNets). We first derive an upper bound of the convergence rate that clearly demonstrates the impact of the model pruning and wireless communications between the clients and the associated BS. Then we jointly optimize the model pruning ratio, central processing unit (CPU) frequency and transmission power of the clients in order to minimize the controllable terms of the convergence bound under strict delay and energy constraints. However, since the original problem is not convex, we perform successive convex approximation (SCA) and jointly optimize the parameters for the relaxed convex problem. Through extensive simulation, we validate the effectiveness of our proposed PHFL algorithm in terms of test accuracy, wall clock time, energy consumption and bandwidth requirement.
comment: Accepted for publications in the IEEE Transactions on Wireless Communications (TWC); \copyright 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses
♻ ☆ 3D Directed Formation Control with Global Shape Convergence using Bispherical Coordinates
In this paper, we present a novel 3D formation control scheme for directed graphs in a leader-follower configuration, achieving (almost) global convergence to the desired shape. Specifically, we introduce three controlled variables representing bispherical coordinates that uniquely describe the formation in 3D. Acyclic triangulated directed graphs (a class of minimally acyclic persistent graphs) are used to model the inter-agent sensing topology, while the agents' dynamics are governed by single-integrator model. Our analysis demonstrates that the proposed decentralized formation controller ensures (almost) global asymptotic stability while avoiding potential shape ambiguities in the final formation. Furthermore, the control laws are implementable in arbitrarily oriented local coordinate frames of follower agents using only low-cost onboard vision sensors, making it suitable for practical applications. Finally, we validate our formation control approach by a simulation study.
comment: Submitted to ECC 2024
♻ ☆ A cGAN Ensemble-based Uncertainty-aware Surrogate Model for Offline Model-based Optimization in Industrial Control Problems IJCNN 2024
This study focuses on two important problems related to applying offline model-based optimization to real-world industrial control problems. The first problem is how to create a reliable probabilistic model that accurately captures the dynamics present in noisy industrial data. The second problem is how to reliably optimize control parameters without actively collecting feedback from industrial systems. Specifically, we introduce a novel cGAN ensemble-based uncertainty-aware surrogate model for reliable offline model-based optimization in industrial control problems. The effectiveness of the proposed method is demonstrated through extensive experiments conducted on two representative cases, namely a discrete control case and a continuous control case. The results of these experiments show that our method outperforms several competitive baselines in the field of offline model-based optimization for industrial control.
comment: Accepted in IJCNN 2024
Robotics 26
☆ Risk-Calibrated Human-Robot Interaction via Set-Valued Intent Prediction
Tasks where robots must cooperate with humans, such as navigating around a cluttered home or sorting everyday items, are challenging because they exhibit a wide range of valid actions that lead to similar outcomes. Moreover, zero-shot cooperation between human-robot partners is an especially challenging problem because it requires the robot to infer and adapt on the fly to a latent human intent, which could vary significantly from human to human. Recently, deep learned motion prediction models have shown promising results in predicting human intent but are prone to being confidently incorrect. In this work, we present Risk-Calibrated Interactive Planning (RCIP), which is a framework for measuring and calibrating risk associated with uncertain action selection in human-robot cooperation, with the fundamental idea that the robot should ask for human clarification when the risk associated with the uncertainty in the human's intent cannot be controlled. RCIP builds on the theory of set-valued risk calibration to provide a finite-sample statistical guarantee on the cumulative loss incurred by the robot while minimizing the cost of human clarification in complex multi-step settings. Our main insight is to frame the risk control problem as a sequence-level multi-hypothesis testing problem, allowing efficient calibration using a low-dimensional parameter that controls a pre-trained risk-aware policy. Experiments across a variety of simulated and real-world environments demonstrate RCIP's ability to predict and adapt to a diverse set of dynamic human intents.
comment: Website with additional information, videos, and code: https://risk-calibrated-planning.github.io/
☆ Explore until Confident: Efficient Exploration for Embodied Question Answering
We consider the problem of Embodied Question Answering (EQA), which refers to settings where an embodied agent such as a robot needs to actively explore an environment to gather information until it is confident about the answer to a question. In this work, we leverage the strong semantic reasoning capabilities of large vision-language models (VLMs) to efficiently explore and answer such questions. However, there are two main challenges when using VLMs in EQA: they do not have an internal memory for mapping the scene to be able to plan how to explore over time, and their confidence can be miscalibrated and can cause the robot to prematurely stop exploration or over-explore. We propose a method that first builds a semantic map of the scene based on depth information and via visual prompting of a VLM - leveraging its vast knowledge of relevant regions of the scene for exploration. Next, we use conformal prediction to calibrate the VLM's question answering confidence, allowing the robot to know when to stop exploration - leading to a more calibrated and efficient exploration strategy. To test our framework in simulation, we also contribute a new EQA dataset with diverse, realistic human-robot scenarios and scenes built upon the Habitat-Matterport 3D Research Dataset (HM3D). Both simulated and real robot experiments show our proposed approach improves the performance and efficiency over baselines that do no leverage VLM for exploration or do not calibrate its confidence. Webpage with experiment videos and code: https://explore-eqa.github.io/
comment: Under review
☆ iA$^*$: Imperative Learning-based A$^*$ Search for Pathfinding
The pathfinding problem, which aims to identify a collision-free path between two points, is crucial for many applications, such as robot navigation and autonomous driving. Classic methods, such as A$^*$ search, perform well on small-scale maps but face difficulties scaling up. Conversely, data-driven approaches can improve pathfinding efficiency but require extensive data labeling and lack theoretical guarantees, making it challenging for practical applications. To combine the strengths of the two methods, we utilize the imperative learning (IL) strategy and propose a novel self-supervised pathfinding framework, termed imperative learning-based A$^*$ (iA$^*$). Specifically, iA$^*$ is a bilevel optimization process where the lower-level optimization is dedicated to finding the optimal path by a differentiable A$^*$ search module, and the upper-level optimization narrows down the search space to improve efficiency via setting suitable initial values from a data-driven model. Besides, the model within the upper-level optimization is a fully convolutional network, trained by the calculated loss in the lower-level optimization. Thus, the framework avoids extensive data labeling and can be applied in diverse environments. Our comprehensive experiments demonstrate that iA$^*$ surpasses both classical and data-driven methods in pathfinding efficiency and shows superior robustness among different tasks, validated with public datasets and simulation environments.
☆ Automated System-level Testing of Unmanned Aerial Systems
Unmanned aerial systems (UAS) rely on various avionics systems that are safety-critical and mission-critical. A major requirement of international safety standards is to perform rigorous system-level testing of avionics software systems. The current industrial practice is to manually create test scenarios, manually/automatically execute these scenarios using simulators, and manually evaluate outcomes. The test scenarios typically consist of setting certain flight or environment conditions and testing the system under test in these settings. The state-of-the-art approaches for this purpose also require manual test scenario development and evaluation. In this paper, we propose a novel approach to automate the system-level testing of the UAS. The proposed approach (AITester) utilizes model-based testing and artificial intelligence (AI) techniques to automatically generate, execute, and evaluate various test scenarios. The test scenarios are generated on the fly, i.e., during test execution based on the environmental context at runtime. The approach is supported by a toolset. We empirically evaluate the proposed approach on two core components of UAS, an autopilot system of an unmanned aerial vehicle (UAV) and cockpit display systems (CDS) of the ground control station (GCS). The results show that the AITester effectively generates test scenarios causing deviations from the expected behavior of the UAV autopilot and reveals potential flaws in the GCS-CDS.
☆ ARO: Large Language Model Supervised Robotics Text2Skill Autonomous Learning
Robotics learning highly relies on human expertise and efforts, such as demonstrations, design of reward functions in reinforcement learning, performance evaluation using human feedback, etc. However, reliance on human assistance can lead to expensive learning costs and make skill learning difficult to scale. In this work, we introduce the Large Language Model Supervised Robotics Text2Skill Autonomous Learning (ARO) framework, which aims to replace human participation in the robot skill learning process with large-scale language models that incorporate reward function design and performance evaluation. We provide evidence that our approach enables fully autonomous robot skill learning, capable of completing partial tasks without human intervention. Furthermore, we also analyze the limitations of this approach in task understanding and optimization stability.
comment: 6 pages, 2 figures
☆ Scaling Learning based Policy Optimization for Temporal Tasks via Dropout
This paper introduces a model-based approach for training feedback controllers for an autonomous agent operating in a highly nonlinear environment. We desire the trained policy to ensure that the agent satisfies specific task objectives, expressed in discrete-time Signal Temporal Logic (DT-STL). One advantage for reformulation of a task via formal frameworks, like DT-STL, is that it permits quantitative satisfaction semantics. In other words, given a trajectory and a DT-STL formula, we can compute the robustness, which can be interpreted as an approximate signed distance between the trajectory and the set of trajectories satisfying the formula. We utilize feedback controllers, and we assume a feed forward neural network for learning these feedback controllers. We show how this learning problem is similar to training recurrent neural networks (RNNs), where the number of recurrent units is proportional to the temporal horizon of the agent's task objectives. This poses a challenge: RNNs are susceptible to vanishing and exploding gradients, and na\"{i}ve gradient descent-based strategies to solve long-horizon task objectives thus suffer from the same problems. To tackle this challenge, we introduce a novel gradient approximation algorithm based on the idea of dropout or gradient sampling. We show that, the existing smooth semantics for robustness are inefficient regarding gradient computation when the specification becomes complex. To address this challenge, we propose a new smooth semantics for DT-STL that under-approximates the robustness value and scales well for backpropagation over a complex specification. We show that our control synthesis methodology, can be quite helpful for stochastic gradient descent to converge with less numerical issues, enabling scalable backpropagation over long time horizons and trajectories over high dimensional state spaces.
☆ Learning Early Social Maneuvers for Enhanced Social Navigation ICRA 2024
Socially compliant navigation is an integral part of safety features in Human-Robot Interaction. Traditional approaches to mobile navigation prioritize physical aspects, such as efficiency, but social behaviors gain traction as robots appear more in daily life. Recent techniques to improve the social compliance of navigation often rely on predefined features or reward functions, introducing assumptions about social human behavior. To address this limitation, we propose a novel Learning from Demonstration (LfD) framework for social navigation that exclusively utilizes raw sensory data. Additionally, the proposed system contains mechanisms to consider the future paths of the surrounding pedestrians, acknowledging the temporal aspect of the problem. The final product is expected to reduce the anxiety of people sharing their environment with a mobile robot, helping them trust that the robot is aware of their presence and will not harm them. As the framework is currently being developed, we outline its components, present experimental results, and discuss future work towards realizing this framework.
comment: Submitted to the workshop of Robot Trust for Symbiotic Societies (RTSS) at ICRA 2024 on March 23, 2024
☆ The Impact of Evolutionary Computation on Robotic Design: A Case Study with an Underactuated Hand Exoskeleton ICRA
Robotic exoskeletons can enhance human strength and aid people with physical disabilities. However, designing them to ensure safety and optimal performance presents significant challenges. Developing exoskeletons should incorporate specific optimization algorithms to find the best design. This study investigates the potential of Evolutionary Computation (EC) methods in robotic design optimization, with an underactuated hand exoskeleton (U-HEx) used as a case study. We propose improving the performance and usability of the U-HEx design, which was initially optimized using a naive brute-force approach, by integrating EC techniques such as Genetic Algorithm and Big Bang-Big Crunch Algorithm. Comparative analysis revealed that EC methods consistently yield more precise and optimal solutions than brute force in a significantly shorter time. This allowed us to improve the optimization by increasing the number of variables in the design, which was impossible with naive methods. The results show significant improvements in terms of the torque magnitude the device transfers to the user, enhancing its efficiency. These findings underline the importance of performing proper optimization while designing exoskeletons, as well as providing a significant improvement to this specific robotic design.
comment: 6 pages (+ref), 4 figures, IEEE International Conference on Robotics and Automation (ICRA) 2024
☆ AirCrab: A Hybrid Aerial-Ground Manipulator with An Active Wheel
Inspired by the behavior of birds, we present AirCrab, a hybrid aerial ground manipulator (HAGM) with a single active wheel and a 3-degree of freedom (3-DoF) manipulator. AirCrab leverages a single point of contact with the ground to reduce position drift and improve manipulation accuracy. The single active wheel enables locomotion on narrow surfaces without adding significant weight to the robot. To realize accurate attitude maintenance using propellers on the ground, we design a control allocation method for AirCrab that prioritizes attitude control and dynamically adjusts the thrust input to reduce energy consumption. Experiments verify the effectiveness of the proposed control method and the gain in manipulation accuracy with ground contact. A series of operations to complete the letters 'NTU' demonstrates the capability of the robot to perform challenging hybrid aerial-ground manipulation missions.
☆ Vid2Real HRI: Align video-based HRI study designs with real-world settings
HRI research using autonomous robots in real-world settings can produce results with the highest ecological validity of any study modality, but many difficulties limit such studies' feasibility and effectiveness. We propose Vid2Real HRI, a research framework to maximize real-world insights offered by video-based studies. The Vid2Real HRI framework was used to design an online study using first-person videos of robots as real-world encounter surrogates. The online study ($n = 385$) distinguished the within-subjects effects of four robot behavioral conditions on perceived social intelligence and human willingness to help the robot enter an exterior door. A real-world, between-subjects replication ($n = 26$) using two conditions confirmed the validity of the online study's findings and the sufficiency of the participant recruitment target ($22$) based on a power analysis of online study results. The Vid2Real HRI framework offers HRI researchers a principled way to take advantage of the efficiency of video-based study modalities while generating directly transferable knowledge of real-world HRI. Code and data from the study are provided at https://vid2real.github.io/vid2realHRI
☆ DriveEnv-NeRF: Exploration of A NeRF-Based Autonomous Driving Environment for Real-World Performance Validation
In this study, we introduce the DriveEnv-NeRF framework, which leverages Neural Radiance Fields (NeRF) to enable the validation and faithful forecasting of the efficacy of autonomous driving agents in a targeted real-world scene. Standard simulator-based rendering often fails to accurately reflect real-world performance due to the sim-to-real gap, which represents the disparity between virtual simulations and real-world conditions. To mitigate this gap, we propose a workflow for building a high-fidelity simulation environment of the targeted real-world scene using NeRF. This approach is capable of rendering realistic images from novel viewpoints and constructing 3D meshes for emulating collisions. The validation of these capabilities through the comparison of success rates in both simulated and real environments demonstrates the benefits of using DriveEnv-NeRF as a real-world performance indicator. Furthermore, the DriveEnv-NeRF framework can serve as a training environment for autonomous driving agents under various lighting conditions. This approach enhances the robustness of the agents and reduces performance degradation when deployed to the target real scene, compared to agents fully trained using the standard simulator rendering pipeline.
comment: Project page: https://github.com/muyishen2040/DriveEnvNeRF
☆ RicMonk: A Three-Link Brachiation Robot with Passive Grippers for Energy-Efficient Brachiation
This paper presents the design, analysis, and performance evaluation of RicMonk, a novel three-link brachiation robot equipped with passive hook-shaped grippers. Brachiation, an agile and energy-efficient mode of locomotion observed in primates, has inspired the development of RicMonk to explore versatile locomotion and maneuvers on ladder-like structures. The robot's anatomical resemblance to gibbons and the integration of a tail mechanism for energy injection contribute to its unique capabilities. The paper discusses the use of the Direct Collocation methodology for optimizing trajectories for the robot's dynamic behaviors and stabilization of these trajectories using a Time-varying Linear Quadratic Regulator. With RicMonk we demonstrate bidirectional brachiation, and provide comparative analysis with its predecessor, AcroMonk - a two-link brachiation robot, to demonstrate that the presence of a passive tail helps improve energy efficiency. The system design, controllers, and software implementation are publicly available on GitHub and the video demonstration of the experiments can be viewed YouTube.
comment: Open sourced system design, controllers, software implementation can be found at https://github.com/dfki-ric-underactuated-lab/ricmonk and a video demonstrating the experiments performed with RicMonk can be found at https://www.youtube.com/watch?v=hOuDQI7CD8w
☆ Distributed Robust Learning based Formation Control of Mobile Robots based on Bioinspired Neural Dynamics
This paper addresses the challenges of distributed formation control in multiple mobile robots, introducing a novel approach that enhances real-world practicability. We first introduce a distributed estimator using a variable structure and cascaded design technique, eliminating the need for derivative information to improve the real time performance. Then, a kinematic tracking control method is developed utilizing a bioinspired neural dynamic-based approach aimed at providing smooth control inputs and effectively resolving the speed jump issue. Furthermore, to address the challenges for robots operating with completely unknown dynamics and disturbances, a learning-based robust dynamic controller is developed. This controller provides real time parameter estimates while maintaining its robustness against disturbances. The overall stability of the proposed method is proved with rigorous mathematical analysis. At last, multiple comprehensive simulation studies have shown the advantages and effectiveness of the proposed method.
comment: This paper is accepted by IEEE Transactions on Intelligent Vehicles
☆ PNAS-MOT: Multi-Modal Object Tracking with Pareto Neural Architecture Search IEEE Robotics and Automation Letters 2024
Multiple object tracking is a critical task in autonomous driving. Existing works primarily focus on the heuristic design of neural networks to obtain high accuracy. As tracking accuracy improves, however, neural networks become increasingly complex, posing challenges for their practical application in real driving scenarios due to the high level of latency. In this paper, we explore the use of the neural architecture search (NAS) methods to search for efficient architectures for tracking, aiming for low real-time latency while maintaining relatively high accuracy. Another challenge for object tracking is the unreliability of a single sensor, therefore, we propose a multi-modal framework to improve the robustness. Experiments demonstrate that our algorithm can run on edge devices within lower latency constraints, thus greatly reducing the computational requirements for multi-modal object tracking while keeping lower latency.
comment: IEEE Robotics and Automation Letters 2024. Code is available at https://github.com/PholyPeng/PNAS-MOT
☆ Data-Driven Predictive Control for Robust Exoskeleton Locomotion
Exoskeleton locomotion must be robust while being adaptive to different users with and without payloads. To address these challenges, this work introduces a data-driven predictive control (DDPC) framework to synthesize walking gaits for lower-body exoskeletons, employing Hankel matrices and a state transition matrix for its data-driven model. The proposed approach leverages DDPC through a multi-layer architecture. At the top layer, DDPC serves as a planner employing Hankel matrices and a state transition matrix to generate a data-driven model that can learn and adapt to varying users and payloads. At the lower layer, our method incorporates inverse kinematics and passivity-based control to map the planned trajectory from DDPC into the full-order states of the lower-body exoskeleton. We validate the effectiveness of this approach through numerical simulations and hardware experiments conducted on the Atalante lower-body exoskeleton with different payloads. Moreover, we conducted a comparative analysis against the model predictive control (MPC) framework based on the reduced-order linear inverted pendulum (LIP) model. Through this comparison, the paper demonstrates that DDPC enables robust bipedal walking at various velocities while accounting for model uncertainties and unknown perturbations.
♻ ☆ LONER: LiDAR Only Neural Representations for Real-Time SLAM
This paper proposes LONER, the first real-time LiDAR SLAM algorithm that uses a neural implicit scene representation. Existing implicit mapping methods for LiDAR show promising results in large-scale reconstruction, but either require groundtruth poses or run slower than real-time. In contrast, LONER uses LiDAR data to train an MLP to estimate a dense map in real-time, while simultaneously estimating the trajectory of the sensor. To achieve real-time performance, this paper proposes a novel information-theoretic loss function that accounts for the fact that different regions of the map may be learned to varying degrees throughout online training. The proposed method is evaluated qualitatively and quantitatively on two open-source datasets. This evaluation illustrates that the proposed loss function converges faster and leads to more accurate geometry reconstruction than other loss functions used in depth-supervised neural implicit frameworks. Finally, this paper shows that LONER estimates trajectories competitively with state-of-the-art LiDAR SLAM methods, while also producing dense maps competitive with existing real-time implicit mapping methods that use groundtruth poses.
comment: First two authors equally contributed. Webpage: https://umautobots.github.io/loner
♻ ☆ ARTEMIS: AI-driven Robotic Triage Labeling and Emergency Medical Information System
Mass casualty incidents (MCIs) pose a significant challenge to emergency medical services by overwhelming available resources and personnel. Effective victim assessment is the key to minimizing casualties during such a crisis. We introduce ARTEMIS, an AI-driven Robotic Triage Labeling and Emergency Medical Information System, to aid first responders in MCI events. It leverages speech processing, natural language processing, and deep learning to help with acuity classification. This is deployed on a quadruped that performs victim localization and preliminary injury severity assessment. First responders access victim information through a Graphical User Interface that is updated in real-time. To validate our proposed algorithmic triage protocol, we used the Unitree Go1 quadruped. The robot identifies humans, interacts with them, gets vitals and information, and assigns an acuity label. Simulations of an MCI in software and a controlled environment outdoors were conducted. The system achieved a triage-level classification precision of over 74% on average and 99% for the most critical victims, i.e. level 1 acuity, outperforming state-of-the-art deep learning-based triage labeling systems. In this paper, we showcase the potential of human-robot interaction in assisting medical personnel in MCI events.
♻ ☆ NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigation
Vision-and-Language Navigation (VLN) stands as a key research problem of Embodied AI, aiming at enabling agents to navigate in unseen environments following linguistic instructions. In this field, generalization is a long-standing challenge, either to out-of-distribution scenes or from Sim to Real. In this paper, we propose NaVid, a video-based large vision language model (VLM), to mitigate such a generalization gap. NaVid makes the first endeavour to showcase the capability of VLMs to achieve state-of-the-art level navigation performance without any maps, odometer and depth inputs. Following human instruction, NaVid only requires an on-the-fly video stream from a monocular RGB camera equipped on the robot to output the next-step action. Our formulation mimics how humans navigate and naturally gets rid of the problems introduced by odometer noises, and the Sim2Real gaps from map or depth inputs. Moreover, our video-based approach can effectively encode the historical observations of robots as spatio-temporal contexts for decision-making and instruction following. We train NaVid with 550k navigation samples collected from VLN-CE trajectories, including action-planning and instruction-reasoning samples, along with 665k large-scale web data. Extensive experiments show that NaVid achieves SOTA performance in simulation environments and the real world, demonstrating superior cross-dataset and Sim2Real transfer. We thus believe our proposed VLM approach plans the next step for not only the navigation agents but also this research field.
♻ ☆ Tactile Estimation of Extrinsic Contact Patch for Stable Placement ICRA2024
Precise perception of contact interactions is essential for fine-grained manipulation skills for robots. In this paper, we present the design of feedback skills for robots that must learn to stack complex-shaped objects on top of each other (see Fig.1). To design such a system, a robot should be able to reason about the stability of placement from very gentle contact interactions. Our results demonstrate that it is possible to infer the stability of object placement based on tactile readings during contact formation between the object and its environment. In particular, we estimate the contact patch between a grasped object and its environment using force and tactile observations to estimate the stability of the object during a contact formation. The contact patch could be used to estimate the stability of the object upon release of the grasp. The proposed method is demonstrated in various pairs of objects that are used in a very popular board game.
comment: Accepted at ICRA2024
♻ ☆ Integration of Large Language Models within Cognitive Architectures for Autonomous Robots IROS 2024
Symbolic reasoning systems have been used in cognitive architectures to provide inference and planning capabilities. However, defining domains and problems has proven difficult and prone to errors. Moreover, Large Language Models (LLMs) have emerged as tools to process natural language for different tasks. In this paper, we propose the use of LLMs to tackle these problems. This way, this paper proposes the integration of LLMs in the ROS 2-integrated cognitive architecture MERLIN2 for autonomous robots. Specifically, we present the design, development and deployment of how to leverage the reasoning capabilities of LLMs inside the deliberative processes of MERLIN2. As a result, the deliberative system is updated from a PDDL-based planner system to a natural language planning system. This proposal is evaluated quantitatively and qualitatively, measuring the impact of incorporating the LLMs in the cognitive architecture. Results show that a classical approach achieves better performance but the proposed solution provides an enhanced interaction through natural language.
comment: 8 pages, 6 figures, 2 tables, Submitted to IROS 2024
♻ ☆ Exploring 3D Human Pose Estimation and Forecasting from the Robot's Perspective: The HARPER Dataset
We introduce HARPER, a novel dataset for 3D body pose estimation and forecast in dyadic interactions between users and Spot, the quadruped robot manufactured by Boston Dynamics. The key-novelty is the focus on the robot's perspective, i.e., on the data captured by the robot's sensors. These make 3D body pose analysis challenging because being close to the ground captures humans only partially. The scenario underlying HARPER includes 15 actions, of which 10 involve physical contact between the robot and users. The Corpus contains not only the recordings of the built-in stereo cameras of Spot, but also those of a 6-camera OptiTrack system (all recordings are synchronized). This leads to ground-truth skeletal representations with a precision lower than a millimeter. In addition, the Corpus includes reproducible benchmarks on 3D Human Pose Estimation, Human Pose Forecasting, and Collision Prediction, all based on publicly available baseline approaches. This enables future HARPER users to rigorously compare their results with those we provide in this work.
♻ ☆ Space Filling Curves for Coverage Path Planning with Online Obstacle Avoidance
The paper presents a strategy for robotic exploration problem using Space-Filling curves (SFC). The strategy plans a path that avoids unknown obstacles while ensuring complete coverage of the free space in region of interest. The region of interest is first tessellated, and the tiles/cells are connected using a SFC pattern. A robot follows the SFC to explore the entire area. However, obstacles can block the systematic movement of the robot. We overcome this problem by determining an alternate path online that avoids the blocked cells while ensuring all the accessible cells are visited at least once. The proposed strategy chooses next waypoint based on the graph connectivity of the cells and the obstacle encountered so far. It is online, exhaustive and works in situations demanding non-uniform coverage. The completeness of the strategy is proved and its desirable properties are discussed with examples.
♻ ☆ Fully Spiking Neural Network for Legged Robots
In recent years, legged robots based on deep reinforcement learning have made remarkable progress. Quadruped robots have demonstrated the ability to complete challenging tasks in complex environments and have been deployed in real-world scenarios to assist humans. Simultaneously, bipedal and humanoid robots have achieved breakthroughs in various demanding tasks. Current reinforcement learning methods can utilize diverse robot bodies and historical information to perform actions. However, prior research has not emphasized the speed and energy consumption of network inference, as well as the biological significance of the neural networks themselves. Most of the networks employed are traditional artificial neural networks that utilize multilayer perceptrons (MLP). In this paper, we successfully apply a novel Spiking Neural Network (SNN) to process legged robots, achieving outstanding results across a range of simulated terrains. SNN holds a natural advantage over traditional neural networks in terms of inference speed and energy consumption, and their pulse-form processing of body perception signals offers improved biological interpretability. Applying more biomimetic neural networks to legged robots can further reduce the heat dissipation and structural burden caused by the high power consumption of neural networks. To the best of our knowledge, this is the first work to implement SNN in legged robots.
♻ ☆ SMART-LLM: Smart Multi-Agent Robot Task Planning using Large Language Models IROS 2024
In this work, we introduce SMART-LLM, an innovative framework designed for embodied multi-robot task planning. SMART-LLM: Smart Multi-Agent Robot Task Planning using Large Language Models (LLMs), harnesses the power of LLMs to convert high-level task instructions provided as input into a multi-robot task plan. It accomplishes this by executing a series of stages, including task decomposition, coalition formation, and task allocation, all guided by programmatic LLM prompts within the few-shot prompting paradigm. We create a benchmark dataset designed for validating the multi-robot task planning problem, encompassing four distinct categories of high-level instructions that vary in task complexity. Our evaluation experiments span both simulation and real-world scenarios, demonstrating that the proposed model can achieve promising results for generating multi-robot task plans. The experimental videos, code, and datasets from the work can be found at https://sites.google.com/view/smart-llm/.
comment: Submitted to IROS 2024
♻ ☆ SurgicalPart-SAM: Part-to-Whole Collaborative Prompting for Surgical Instrument Segmentation
The Segment Anything Model (SAM) exhibits promise in generic object segmentation and offers potential for various applications. Existing methods have applied SAM to surgical instrument segmentation (SIS) by tuning SAM-based frameworks with surgical data. However, they fall short in two crucial aspects: (1) Straightforward model tuning with instrument masks treats each instrument as a single entity, neglecting their complex structures and fine-grained details; and (2) Instrument category-based prompts are not flexible and informative enough to describe instrument structures. To address these problems, in this paper, we investigate text promptable SIS and propose SurgicalPart-SAM (SP-SAM), a novel SAM efficient-tuning approach that explicitly integrates instrument structure knowledge with SAM's generic knowledge, guided by expert knowledge on instrument part compositions. Specifically, we achieve this by proposing (1) Collaborative Prompts that describe instrument structures via collaborating category-level and part-level texts; (2) Cross-Modal Prompt Encoder that encodes text prompts jointly with visual embeddings into discriminative part-level representations; and (3) Part-to-Whole Adaptive Fusion and Hierarchical Decoding that adaptively fuse the part-level representations into a whole for accurate instrument segmentation in surgical scenarios. Built upon them, SP-SAM acquires a better capability to comprehend surgical instruments in terms of both overall structure and part-level details. Extensive experiments on both the EndoVis2018 and EndoVis2017 datasets demonstrate SP-SAM's state-of-the-art performance with minimal tunable parameters. The code will be available at https://github.com/wenxi-yue/SurgicalPart-SAM.
comment: Technical Report. The source code will be released at https://github.com/wenxi-yue/SurgicalPart-SAM
♻ ☆ Bird's Eye View Based Pretrained World model for Visual Navigation IROS 2024
Sim2Real transfer has gained popularity because it helps transfer from inexpensive simulators to real world. This paper presents a novel system that fuses components in a traditional World Model into a robust system, trained entirely within a simulator, that Zero-Shot transfers to the real world. To facilitate transfer, we use an intermediary representation that is based on \textit{Bird's Eye View (BEV)} images. Thus, our robot learns to navigate in a simulator by first learning to translate from complex \textit{First-Person View (FPV)} based RGB images to BEV representations, then learning to navigate using those representations. Later, when tested in the real world, the robot uses the perception model that translates FPV-based RGB images to embeddings that were learned by the FPV to BEV translator and that can be used by the downstream policy. The incorporation of state-checking modules using \textit{Anchor images} and Mixture Density LSTM not only interpolates uncertain and missing observations but also enhances the robustness of the model in the real-world. We trained the model using data from a Differential drive robot in the CARLA simulator. Our methodology's effectiveness is shown through the deployment of trained models onto a real-world Differential drive robot. Lastly we release a comprehensive codebase, dataset and models for training and deployment (\url{https://sites.google.com/view/value-explicit-pretraining}).
comment: Under Review at the IROS 2024; Accepted at NeurIPS 2023, Robot Learning Workshop
Artificial Intelligence 6
☆ Detection of Problem Gambling with Less Features Using Machine Learning Methods
Analytic features in gambling study are performed based on the amount of data monitoring on user daily actions. While performing the detection of problem gambling, existing datasets provide relatively rich analytic features for building machine learning based model. However, considering the complexity and cost of collecting the analytic features in real applications, conducting precise detection with less features will tremendously reduce the cost of data collection. In this study, we propose a deep neural networks PGN4 that performs well when using limited analytic features. Through the experiment on two datasets, we discover that PGN4 only experiences a mere performance drop when cutting 102 features to 5 features. Besides, we find the commonality within the top 5 features from two datasets.
comment: 6 pages, 5 tables, 1 figure
☆ SAT Encoding of Partial Ordering Models for Graph Coloring Problems
In this paper, we suggest new SAT encodings of the partial-ordering based ILP model for the graph coloring problem (GCP) and the bandwidth coloring problem (BCP). The GCP asks for the minimum number of colors that can be assigned to the vertices of a given graph such that each two adjacent vertices get different colors. The BCP is a generalization, where each edge has a weight that enforces a minimal "distance" between the assigned colors, and the goal is to minimize the "largest" color used. For the widely studied GCP, we experimentally compare our new SAT encoding to the state-of-the-art approaches on the DIMACS benchmark set. Our evaluation confirms that this SAT encoding is effective for sparse graphs and even outperforms the state-of-the-art on some DIMACS instances. For the BCP, our theoretical analysis shows that the partial-ordering based SAT and ILP formulations have an asymptotically smaller size than that of the classical assignment-based model. Our practical evaluation confirms not only a dominance compared to the assignment-based encodings but also to the state-of-the-art approaches on a set of benchmark instances. Up to our knowledge, we have solved several open instances of the BCP from the literature for the first time.
☆ Finding needles in a haystack: A Black-Box Approach to Invisible Watermark Detection
In this paper, we propose WaterMark Detection (WMD), the first invisible watermark detection method under a black-box and annotation-free setting. WMD is capable of detecting arbitrary watermarks within a given reference dataset using a clean non-watermarked dataset as a reference, without relying on specific decoding methods or prior knowledge of the watermarking techniques. We develop WMD using foundations of offset learning, where a clean non-watermarked dataset enables us to isolate the influence of only watermarked samples in the reference dataset. Our comprehensive evaluations demonstrate the effectiveness of WMD, significantly outperforming naive detection methods, which only yield AUC scores around 0.5. In contrast, WMD consistently achieves impressive detection AUC scores, surpassing 0.9 in most single-watermark datasets and exceeding 0.7 in more challenging multi-watermark scenarios across diverse datasets and watermarking methods. As invisible watermarks become increasingly prevalent, while specific decoding techniques remain undisclosed, our approach provides a versatile solution and establishes a path toward increasing accountability, transparency, and trust in our digital visual content.
☆ Understanding The Effectiveness of Lossy Compression in Machine Learning Training Sets
Learning and Artificial Intelligence (ML/AI) techniques have become increasingly prevalent in high performance computing (HPC). However, these methods depend on vast volumes of floating point data for training and validation which need methods to share the data on a wide area network (WAN) or to transfer it from edge devices to data centers. Data compression can be a solution to these problems, but an in-depth understanding of how lossy compression affects model quality is needed. Prior work largely considers a single application or compression method. We designed a systematic methodology for evaluating data reduction techniques for ML/AI, and we use it to perform a very comprehensive evaluation with 17 data reduction methods on 7 ML/AI applications to show modern lossy compression methods can achieve a 50-100x compression ratio improvement for a 1% or less loss in quality. We identify critical insights that guide the future use and design of lossy compressors for ML/AI.
comment: 12 pages, 4 figures
☆ Adaptive Super Resolution For One-Shot Talking-Head Generation
The one-shot talking-head generation learns to synthesize a talking-head video with one source portrait image under the driving of same or different identity video. Usually these methods require plane-based pixel transformations via Jacobin matrices or facial image warps for novel poses generation. The constraints of using a single image source and pixel displacements often compromise the clarity of the synthesized images. Some methods try to improve the quality of synthesized videos by introducing additional super-resolution modules, but this will undoubtedly increase computational consumption and destroy the original data distribution. In this work, we propose an adaptive high-quality talking-head video generation method, which synthesizes high-resolution video without additional pre-trained modules. Specifically, inspired by existing super-resolution methods, we down-sample the one-shot source image, and then adaptively reconstruct high-frequency details via an encoder-decoder module, resulting in enhanced video clarity. Our method consistently improves the quality of generated videos through a straightforward yet effective strategy, substantiated by quantitative and qualitative evaluations. The code and demo video are available on: \url{https://github.com/Songluchuan/AdaSR-TalkingHead/}.
comment: 5 pages, 3 figures
♻ ☆ Learning Concepts Definable in First-Order Logic with Counting
We study Boolean classification problems over relational background structures in the logical framework introduced by Grohe and Tur\'an (TOCS 2004). It is known (Grohe and Ritzert, LICS 2017) that classifiers definable in first-order logic over structures of polylogarithmic degree can be learned in sublinear time, where the degree of the structure and the running time are measured in terms of the size of the structure. We generalise the results to the first-order logic with counting FOCN, which was introduced by Kuske and Schweikardt (LICS 2017) as an expressive logic generalising various other counting logics. Specifically, we prove that classifiers definable in FOCN over classes of structures of polylogarithmic degree can be consistently learned in sublinear time. This can be seen as a first step towards extending the learning framework to include numerical aspects of machine learning. We extend the result to agnostic probably approximately correct (PAC) learning for classes of structures of degree at most $(\log \log n)^c$ for some constant $c$. Moreover, we show that bounding the degree is crucial to obtain sublinear-time learning algorithms. That is, we prove that, for structures of unbounded degree, learning is not possible in sublinear time, even for classifiers definable in plain first-order logic.
Optimization and Control 24
☆ Risk-Calibrated Human-Robot Interaction via Set-Valued Intent Prediction
Tasks where robots must cooperate with humans, such as navigating around a cluttered home or sorting everyday items, are challenging because they exhibit a wide range of valid actions that lead to similar outcomes. Moreover, zero-shot cooperation between human-robot partners is an especially challenging problem because it requires the robot to infer and adapt on the fly to a latent human intent, which could vary significantly from human to human. Recently, deep learned motion prediction models have shown promising results in predicting human intent but are prone to being confidently incorrect. In this work, we present Risk-Calibrated Interactive Planning (RCIP), which is a framework for measuring and calibrating risk associated with uncertain action selection in human-robot cooperation, with the fundamental idea that the robot should ask for human clarification when the risk associated with the uncertainty in the human's intent cannot be controlled. RCIP builds on the theory of set-valued risk calibration to provide a finite-sample statistical guarantee on the cumulative loss incurred by the robot while minimizing the cost of human clarification in complex multi-step settings. Our main insight is to frame the risk control problem as a sequence-level multi-hypothesis testing problem, allowing efficient calibration using a low-dimensional parameter that controls a pre-trained risk-aware policy. Experiments across a variety of simulated and real-world environments demonstrate RCIP's ability to predict and adapt to a diverse set of dynamic human intents.
comment: Website with additional information, videos, and code: https://risk-calibrated-planning.github.io/
☆ Putting all eggs in one basket: some insights from a correlation inequality
We give examples of situations -- stochastic production, military tactics, corporate merger -- where it is beneficial to concentrate risk rather than to diversify it, that is, to put all eggs in one basket. Our examples admit a dual interpretation: as optimal strategies of a single player (the `principal') or, alternatively, as dominant strategies in a non-cooperative game with multiple players (the `agents'). The key mathematical result can be formulated in terms of a convolution structure on the set of increasing functions on a Boolean lattice (the lattice of subsets of a finite set). This generalizes the well-known Harris inequality from statistical physics and discrete mathematics; we give a simple self-contained proof of this result, and prove a further generalization based on the game-theoretic approach.
comment: 16 pages
☆ Michell Truss and From 1-beam to k-beam
This paper generalizes the Michell Truss problem and Gangbo's paper from 1-dimension to higher dimensions using geometric measure theory. Given an elastic surface $S$ made of $(k-1)$-beams under an equilibriated system $F$ of external forces, then we ask the following two questions: 1. What are the necessary and sufficient conditions for the existence of an elastic body made of $k$-beams whose forces on the surface balance $F$ and whose surfaces consist of $S$. 2. What is an optimal design so that the total cost is a minimum? We've solved the existence question completely; and research is still in progress for the minimal question. In particular when $k=1$, it involves a system of beams joining a given finite collection of pointed forces. It was first introduced by A. Michell in 1904, then used in mechanical engineering, and recently popularized in many pure mathematics works by W. Gangbo, Prager, and others. Here we are going to generalize them to higher dimensional cases. We have already found the minimal solutions in terms of the flat chain complex and vector-valued currents. Right now we are studying the Calibration theory for future directions. I appreciate the discussion with Prof. Robert Hardt!
☆ GPU-accelerated nonlinear model predictive control with ExaModels and MadNLP
We investigate the potential of Graphics Processing Units (GPUs) to solve large-scale nonlinear model predictive control (NMPC) problems. We accelerate the solution of the constrained nonlinear programs in the NMPC algorithm using the GPU-accelerated automatic differentiation tool ExaModels with the interior-point solver MadNLP. The sparse linear systems formulated in the interior-point method is solved on the GPU using a hybrid solver combining an iterative method with a sparse Cholesky factorization, which harness the newly released NVIDIA cuDSS solver. Our results on the classical distillation column instance show that despite a significant pre-processing time, the hybrid solver allows to reduce the time per iteration by a factor of 25 for the largest instance.
comment: 6 pages, 1 figure, 2 tables
☆ Semi-on-Demand Hybrid Transit Route Design with Shared Autonomous Mobility Services
This study examines the route design of a semi-on-demand hybrid route directional service in the public transit network, offering on-demand flexible route service in low-density areas and fixed route service in higher-density areas with Shared Autonomous Mobility Service (SAMS). The study develops analytically tractable cost expressions that capture access, waiting, and riding costs for users, and distance-based operating and time-based vehicle costs for operators. Two formulations are presented for strategic and tactical decisions in flexible route portion, fleet size, headway, and vehicle size optimization, enabling the determination of route types between fixed, hybrid, and flexible routes based on demand, cost, and operational parameters. The practical applications and benefits of semi-on-demand feeders are demonstrated with numerical examples and a large-scale case study in the Chicago metropolitan area. Findings reveal scenarios in which flexible route portions serving passengers located further away reduce total costs, particularly user costs. Lower operating costs in lower-demand areas favor more flexible routes, whereas higher demand densities favor more traditional line-based operations. On two studied lines, a current cost forecast favors smaller vehicles with flexible routes, but operating constraints and higher operating costs would favor bigger vehicles with hybrid routes. The study provides an analytical tool to design SAMS as directional services and transit feeders, and tractable continuous approximation formulations for future research in transit network design.
comment: 24 pages, 12 figures, the previous version presented at the 103rd Transportation Research Board Annual Meeting, Washington, D.C
☆ Divergence conforming DG method for the optimal control of the Oseen equation with variable viscosity
This study introduces the divergence-conforming discontinuous Galerkin finite element method (DGFEM) for numerically approximating optimal control problems with distributed constraints, specifically those governed by stationary generalized Oseen equations. We provide optimal a priori error estimates in energy norms for such problems using the divergence-conforming DGFEM approach. Moreover, we thoroughly analyze $L^2$ error estimates for scenarios dominated by diffusion and convection. Additionally, we establish the new reliable and efficient a posteriori error estimators for the optimal control of the Oseen equation with variable viscosity. Theoretical findings are validated through numerical experiments conducted in both two and three dimensions.
☆ Horoballs and the subgradient method
To explore convex optimization on Hadamard spaces, we consider an iteration in the style of a subgradient algorithm. Traditionally, such methods assume that the underlying spaces are manifolds and that the objectives are geodesically convex: the methods are described using tangent spaces and exponential maps. By contrast, our iteration applies in a general Hadamard space, is framed in the underlying space itself, and relies instead on horospherical convexity of the objective level sets. For this restricted class of objectives, we prove a complexity result of the usual form. Notably, the complexity does not depend on a lower bound on the space curvature.
☆ Fast Consensus Topology Design via Minimizing Laplacian Energy
This paper characterizes the graphical properties of an optimal topology with minimal Laplacian energy under the constraint of fixed numbers of vertices and edges, and devises an algorithm to construct such connected optimal graphs. These constructed graphs possess maximum vertex and edge connectivity, and more importantly, exhibit large algebraic connectivity of an optimal order provided they are not sparse. These properties guarantee fast and resilient consensus processes over these graphs.
Optimal Hospital Capacity Management During Demand Surges
Effective hospital capacity management is pivotal for enhancing patient care quality, operational efficiency, and healthcare system resilience, notably during demand spikes like those seen in the COVID-19 pandemic. However, devising optimal capacity strategies is complicated by fluctuating demand, conflicting objectives, and multifaceted practical constraints. This study presents a data-driven framework to optimize capacity management decisions within hospital systems during surge events. Two key decisions are optimized over a tactical planning horizon: allocating dedicated capacity to surge patients and transferring incoming patients between emergency departments (EDs) of hospitals to better distribute demand. The optimization models are formulated as robust mixed-integer linear programs, enabling efficient computation of optimal decisions that are robust against demand uncertainty. The models incorporate practical constraints and costs, including setup times and costs for adding surge capacity, restrictions on ED patient transfers, and relative costs of different decisions that reflect impacts on care quality and operational efficiency. The methodology is evaluated retrospectively in a hospital system during the height of the COVID-19 pandemic to demonstrate the potential impact of the recommended decisions. The results show that optimally allocating beds and transferring just 30 patients over a 63 day period around the peak, less than one transfer every two days, could have reduced the need for surge capacity in the hospital system by approximately 98%. Overall, this work introduces a practical tool to transform capacity management decision-making, enabling proactive planning and the use of data-driven recommendations to improve outcomes.
☆ Properties for transposition solutions to operator-valued BSEEs, and applications to robust second order necessary conditions for controlled SEEs
This article is concerned with the second order necessary conditions for the stochastic optimal control problem of stochastic evolution equation with model uncertainty when the traditional Pontryagin-type maximum principle holds trivially and do not provide any information depicting the optimal control. The diffusion term of the state equation is allowed to be control dependent with convex control constraints. Transposition method is adopted to deal with the adjoint operator-valued backward stochastic evolution equations, especially the correction terms. Besides, weak convergence arguments are performed to obtain the optimal uncertainty measure, among which the regularities of the state processes, variational processes, and adjoint processes (in the transposition sense) are carefully characterized. Malliavin calculus is applied to pave the way for Lebesgue differentiation theorem to deduce the pointwise robust optimality conditions.
comment: 40 pages
☆ Robust pointwise second order necessary conditions for singular stochastic optimal control with model uncertainty
We study the singular stochastic optimal control problem with model uncertainty, where the necessary conditions determined by the corresponding maximum principle are trivial. Robust integral form and pointwise second order necessary optimality conditions under certain compactness conditions are derived. Both the drift and diffusion terms are control dependent but the control region are assumed to be convex. The convex variational method is employed, because linear structure is essential in deriving the weak limit of uncertainty measures. Other main technical ingredients in obtaining the integral type conditions are compact analysis and minimax theorem, while for the pointwise ones it is Clark-Ocone formula and Lebesgue differentiation type theorem. Besides, a compendious example is given to illustrate the motivation and effectiveness of the results.
comment: 36 pages
☆ Causal Tracking of Distributions in Wasserstein Space: A Model Predictive Control Scheme
We consider the problem of optimal swarm tracking which can be formulated as a tracking problem for distributions in the Wasserstein metric. Optimal control solutions to this problem are non-causal and require knowing the time-trajectory of the distribution to be tracked in advance. We propose a scheme where these non-causal solutions can be used together with a predictive model for the reference to achieve causal tracking control of a priori-unknown references. We develop the resulting model-predictive control scheme in the simple case where the reference is predicted to be constant-in-time. A computational algorithm based on particle methods and discrete optimal mass transport is presented, and numerical simulations are provided for various classes of reference signals. The results demonstrate that the proposed control algorithm achieves reasonable performance even when using simple predictive models.
comment: 8 pages, 7 figures
☆ Conservative Surrogate Models for Optimization with the Active Subspace Method
We are interested in building low-dimensional surrogate models to reduce optimization costs, while having theoretical guarantees that the optimum will satisfy the constraints of the full-size model, by making conservative approximations. The surrogate model is constructed using a Gaussian process regression (GPR). To ensure conservativeness, two new approaches are proposed: the first one using bootstrapping, and the second one using concentration inequalities. Those two techniques are based on a stochastic argument and thus will only enforce conservativeness up to a user-defined probability threshold. The method has applications in the context of optimization using the active subspace method for dimensionality reduction of the objective function and the constraints, addressing recorded issues about constraint violations. The resulting algorithms are tested on a toy optimization problem in thermal design.
Review of Large-Scale Simulation Optimization
Large-scale simulation optimization (SO) problems encompass both large-scale ranking-and-selection problems and high-dimensional discrete or continuous SO problems, presenting significant challenges to existing SO theories and algorithms. This paper begins by providing illustrative examples that highlight the differences between large-scale SO problems and those of a more moderate scale. Subsequently, it reviews several widely employed techniques for addressing large-scale SO problems, such as divide and conquer, dimension reduction, and gradient-based algorithms. Additionally, the paper examines parallelization techniques leveraging widely accessible parallel computing environments to facilitate the resolution of large-scale SO problems.
☆ Constraint Preconditioning and Parameter Selection for a First-Order Primal-Dual Method applied to Model Predictive Control
Many techniques for real-time trajectory optimization and control require the solution of optimization problems at high frequencies. However, ill-conditioning in the optimization problem can significantly reduce the speed of first-order primal-dual optimization algorithms. We introduce a preconditioning technique and step-size heuristic for Proportional-Integral Projected Gradient (PIPG), a first-order primal-dual algorithm. The preconditioning technique, based on the QR factorization, aims to reduce the condition number of the KKT matrix associated with the optimization problem. Our step-size selection heuristic chooses step-sizes to minimize the upper bound on the convergence of the primal-dual gap for the optimization problem. These algorithms are tested on two model predictive control problem examples and show a solve-time reduction of at least 3.6x.
☆ The Effectiveness of Local Updates for Decentralized Learning under Data Heterogeneity
We revisit two fundamental decentralized optimization methods, Decentralized Gradient Tracking (DGT) and Decentralized Gradient Descent (DGD), with multiple local updates. We consider two settings and demonstrate that incorporating $K > 1$ local update steps can reduce communication complexity. Specifically, for $\mu$-strongly convex and $L$-smooth loss functions, we proved that local DGT achieves communication complexity $\tilde{\mathcal{O}} \Big(\frac{L}{\mu K} + \frac{\delta}{\mu (1 - \rho)} + \frac{\rho }{(1 - \rho)^2} \cdot \frac{L+ \delta}{\mu}\Big)$, where $\rho$ measures the network connectivity and $\delta$ measures the second-order heterogeneity of the local loss. Our result reveals the tradeoff between communication and computation and shows increasing $K$ can effectively reduce communication costs when the data heterogeneity is low and the network is well-connected. We then consider the over-parameterization regime where the local losses share the same minimums, we proved that employing local updates in DGD, even without gradient correction, can yield a similar effect as DGT in reducing communication complexity. Numerical experiments validate our theoretical results.
☆ Safe Reinforcement Learning for Constrained Markov Decision Processes with Stochastic Stopping Time
In this paper, we present an online reinforcement learning algorithm for constrained Markov decision processes with a safety constraint. Despite the necessary attention of the scientific community, considering stochastic stopping time, the problem of learning optimal policy without violating safety constraints during the learning phase is yet to be addressed. To this end, we propose an algorithm based on linear programming that does not require a process model. We show that the learned policy is safe with high confidence. We also propose a method to compute a safe baseline policy, which is central in developing algorithms that do not violate the safety constraints. Finally, we provide simulation results to show the efficacy of the proposed algorithm. Further, we demonstrate that efficient exploration can be achieved by defining a subset of the state-space called proxy set.
☆ Balancing art and money in pursuit of a Kelly-type optimality
We introduce and study a mathematical model of an art collector. In our model, the collector is a rational agent whose actions in the art market are driven by two competing long-term objectives, namely sustainable financial health and maintaining the collection. Mathematically, our model is a two-dimensional random linear dynamical system with transformation matrix of a peculiar type. In some examples we are able to show that within the Kelly-type optimization paradigm, that is optimizing the system's Lyapunov exponent over a set of policy parameters, the dilemma ``art or money" can be successfully resolved, namely the optimal policy creates a coexistence equilibrium where the value of both is increasing over the time.
♻ ☆ Consensus-Based Optimization Methods Converge Globally
In this paper we study consensus-based optimization (CBO), which is a multi-agent metaheuristic derivative-free optimization method that can globally minimize nonconvex nonsmooth functions and is amenable to theoretical analysis. Based on an experimentally supported intuition that, on average, CBO performs a gradient descent of the squared Euclidean distance to the global minimizer, we devise a novel technique for proving the convergence to the global minimizer in mean-field law for a rich class of objective functions. The result unveils internal mechanisms of CBO that are responsible for the success of the method. In particular, we prove that CBO performs a convexification of a large class of optimization problems as the number of optimizing agents goes to infinity. Furthermore, we improve prior analyses by requiring mild assumptions about the initialization of the method and by covering objectives that are merely locally Lipschitz continuous. As a core component of this analysis, we establish a quantitative nonasymptotic Laplace principle, which may be of independent interest. From the result of CBO convergence in mean-field law, it becomes apparent that the hardness of any global optimization problem is necessarily encoded in the rate of the mean-field approximation, for which we provide a novel probabilistic quantitative estimate. The combination of these results allows to obtain probabilistic global convergence guarantees of the numerical CBO method.
comment: 41 pages, 3 figures
♻ ☆ Policy Evaluation in Distributional LQR (Extended Version)
Distributional reinforcement learning (DRL) enhances the understanding of the effects of the randomness in the environment by letting agents learn the distribution of a random return, rather than its expected value as in standard reinforcement learning. Meanwhile, a challenge in DRL is that the policy evaluation typically relies on the representation of the return distribution, which needs to be carefully designed. In this paper, we address this challenge for the special class of DRL problems that rely on a discounted linear quadratic regulator (LQR), which we call \emph{distributional LQR}. Specifically, we provide a closed-form expression for the distribution of the random return, which is applicable for all types of exogenous disturbance as long as it is independent and identically distributed (i.i.d.). We show that the variance of the random return is bounded if the fourth moment of the exogenous disturbance is bounded. Furthermore, we investigate the sensitivity of the return distribution to model perturbations. While the proposed exact return distribution consists of infinitely many random variables, we show that this distribution can be well approximated by a finite number of random variables. The associated approximation error can be analytically bounded under mild assumptions. When the model is unknown, we propose a model-free approach for estimating the return distribution, supported by sample complexity guarantees. Finally, we extend our approach to partially observable linear systems. Numerical experiments are provided to illustrate the theoretical results.
comment: This draft is an extension to our previous work https://proceedings.mlr.press/v211/wang23c/wang23c.pdf. arXiv admin note: text overlap with arXiv:2303.13657
♻ ☆ Convergence Analysis of Stochastic Gradient Descent with MCMC Estimators
Understanding stochastic gradient descent (SGD) and its variants is essential for machine learning. However, most of the preceding analyses are conducted under amenable conditions such as unbiased gradient estimator and bounded objective functions, which does not encompass many sophisticated applications, such as variational Monte Carlo, entropy-regularized reinforcement learning and variational inference. In this paper, we consider the SGD algorithm that employ the Markov Chain Monte Carlo (MCMC) estimator to compute the gradient, called MCMC-SGD. Since MCMC reduces the sampling complexity significantly, it is an asymptotically convergent biased estimator in practice. Moreover, by incorporating a general class of unbounded functions, it is much more difficult to analyze the MCMC sampling error. Therefore, we assume that the function is sub-exponential and use the Bernstein inequality for non-stationary Markov chains to derive error bounds of the MCMC estimator. Consequently, MCMC-SGD is proven to have a first order convergence rate $O(\log K/\sqrt{n K})$ with $K$ iterations and a sample size $n$. It partially explains how MCMC influences the behavior of SGD. Furthermore, we verify the correlated negative curvature condition under reasonable assumptions. It is shown that MCMC-SGD escapes from saddle points and reaches $(\epsilon,\epsilon^{1/4})$ approximate second order stationary points or $\epsilon^{1/2}$-variance points at least $O(\epsilon^{-11/2}\log^{2}(1/\epsilon) )$ steps with high probability. Our analysis unveils the convergence pattern of MCMC-SGD across a broad class of stochastic optimization problems, and interprets the convergence phenomena observed in practical applications.
♻ ☆ Non-convex potential games for finding global solutions to sensor network localization
Sensor network localization (SNL) problems require determining the physical coordinates of all sensors in a network. This process relies on the global coordinates of anchors and the available measurements between non-anchor and anchor nodes. Attributed to the intrinsic non-convexity, obtaining a globally optimal solution to SNL is challenging, as well as implementing corresponding algorithms. In this paper, we formulate a non-convex multi-player potential game for a generic SNL problem to investigate the identification condition of the global Nash equilibrium (NE) therein, where the global NE represents the global solution of SNL. We employ canonical duality theory to transform the non-convex game into a complementary dual problem. Then we develop a conjugation-based algorithm to compute the stationary points of the complementary dual problem. On this basis, we show an identification condition of the global NE: the stationary point of the proposed algorithm satisfies a duality relation. Finally, simulation results are provided to validate the effectiveness of the theoretical results.
♻ ☆ Performance-Guaranteed Solutions for Multi-Agent Optimal Coverage Problems using Submodularity, Curvature, and Greedy Algorithms
We consider a class of multi-agent optimal coverage problems in which the goal is to determine the optimal placement of a group of agents in a given mission space so that they maximize a coverage objective that represents a blend of individual and collaborative event detection capabilities. This class of problems is extremely challenging due to the non-convex nature of the mission space and of the coverage objective. With this motivation, greedy algorithms are often used as means of getting feasible coverage solutions efficiently. Even though such greedy solutions are suboptimal, the submodularity (diminishing returns) property of the coverage objective can be exploited to provide performance bound guarantees. Moreover, we show that improved performance bound guarantees (beyond the standard (1-1/e) performance bound) can be established using various curvature measures of the coverage problem. In particular, we provide a brief review of all existing popular applicable curvature measures, including a recent curvature measure that we proposed, and discuss their effectiveness and computational complexity, in the context of optimal coverage problems. We also propose novel computationally efficient techniques to estimate some curvature measures. Finally, we provide several numerical results to support our findings and propose several potential future research directions.
comment: Will be submitted to CDC 2024
♻ ☆ Latent assimilation with implicit neural representations for unknown dynamics
Data assimilation is crucial in a wide range of applications, but it often faces challenges such as high computational costs due to data dimensionality and incomplete understanding of underlying mechanisms. To address these challenges, this study presents a novel assimilation framework, termed Latent Assimilation with Implicit Neural Representations (LAINR). By introducing Spherical Implicit Neural Representations (SINR) along with a data-driven uncertainty estimator of the trained neural networks, LAINR enhances efficiency in assimilation process. Experimental results indicate that LAINR holds certain advantage over existing methods based on AutoEncoders, both in terms of accuracy and efficiency.
comment: 40 pages
Systems and Control 26
☆ Risk-Calibrated Human-Robot Interaction via Set-Valued Intent Prediction
Tasks where robots must cooperate with humans, such as navigating around a cluttered home or sorting everyday items, are challenging because they exhibit a wide range of valid actions that lead to similar outcomes. Moreover, zero-shot cooperation between human-robot partners is an especially challenging problem because it requires the robot to infer and adapt on the fly to a latent human intent, which could vary significantly from human to human. Recently, deep learned motion prediction models have shown promising results in predicting human intent but are prone to being confidently incorrect. In this work, we present Risk-Calibrated Interactive Planning (RCIP), which is a framework for measuring and calibrating risk associated with uncertain action selection in human-robot cooperation, with the fundamental idea that the robot should ask for human clarification when the risk associated with the uncertainty in the human's intent cannot be controlled. RCIP builds on the theory of set-valued risk calibration to provide a finite-sample statistical guarantee on the cumulative loss incurred by the robot while minimizing the cost of human clarification in complex multi-step settings. Our main insight is to frame the risk control problem as a sequence-level multi-hypothesis testing problem, allowing efficient calibration using a low-dimensional parameter that controls a pre-trained risk-aware policy. Experiments across a variety of simulated and real-world environments demonstrate RCIP's ability to predict and adapt to a diverse set of dynamic human intents.
comment: Website with additional information, videos, and code: https://risk-calibrated-planning.github.io/
☆ Convection-Enabled Boundary Control of a 2D Channel Flow IEEE Conference on Decision and Control
We consider the incompressible Navier-Stokes equations in a two-dimensional channel. The tangential and normal velocities are assumed to be periodic in the streamwise (horizontal) direction. Moreover, we consider no-slip boundary conditions on the tangential velocity at the top and bottom walls of the channel, and normal velocity actuation at the top and bottom walls. For an arbitrarily large Reynolds number, we design the boundary control inputs to achieve global exponential stabilization, in the L2 sense, of a chosen parabolic Poiseuille profile. Moreover, we design the control inputs such that they have zero mean, but non-zero cubic mean. The zero-mean property is to ensure that the conservation of mass constraint is verified. The non-zero cubic mean property is the key to exploiting the stabilizing effect of nonlinear convection and achieving global stabilization independently of the size of the Reynolds number. This paper is not only the first work where a closed-form feedback law is proposed for global stabilization of parabolic Poiseuille profiles for arbitrary Reynolds number but is also the first generalization of the Cardano-Lyapunov formula, designed initially to stabilize scalar-valued convective PDEs, to a vector-valued convective PDE with a divergence-free constraint on the state.
comment: Submitted to the 63rd IEEE Conference on Decision and Control (CDC 2024)
☆ Perception and Control of Surfing in Virtual Reality using a 6-DoF Motion Platform
The paper presents a system for simulating surfing in Virtual Reality (VR), emphasizing the recreation of aquatic motions and user-initiated propulsive forces using a 6-Degree of Freedom (DoF) motion platform. We present an algorithmic approach to accurately render surfboard kinematics and interactive paddling dynamics, validated through experimental evaluation with \(N=17\) participants. Results indicate that the system effectively reproduces various acceleration levels, the perception of which is independent of users' body posture. We additionally found that the presence of ocean ripples amplifies the perception of acceleration. This system aims to enhance the realism and interactivity of VR surfing, laying a foundation for future advancements in surf therapy and interactive aquatic VR experiences.
☆ From Raw Data to Safety: Reducing Conservatism by Set Expansion
In response to safety concerns associated with learning-based algorithms, safety filters have been proposed as a modular technique. Generally, these filters heavily rely on the system's model, which is contradictory if they are intended to enhance a data-driven or end-to-end learning solution. This paper extends our previous work, a purely Data-Driven Safety Filter (DDSF) based on Willems' lemma, to an extremely short-sighted and non-conservative solution. Specifically, we propose online and offline sample-based methods to expand the safe set of DDSF and reduce its conservatism. Since this method is defined in an input-output framework, it can systematically handle both unknown and time-delay LTI systems using only one single batch of data. To evaluate its performance, we apply the proposed method to a time-delay system under various settings. The simulation results validate the effectiveness of the set expansion algorithm in generating a notably large input-output safe set, resulting in safety filters that are not conservative, even with an extremely short prediction horizon.
☆ A Modular Safety Filter for Safety-Certified Cyber-Physical Systems
Nowadays, many control systems are networked and embed communication and computation capabilities. Such control architectures are prone to cyber attacks on the cyberinfrastructure. Consequently, there is an impellent need to develop solutions to preserve the plant's safety against potential attacks. To ensure safety, this paper introduces a modular safety filter approach that is effective for a variety of cyber-attack types. This solution can be implemented in combination with existing control and detection algorithms, effectively separating safety from performance. The safety filter does not require information on the reliability of the received command or the feature of the used anomaly detector. It can be implemented in conjunction with high-performance, resilient controllers, to achieve both high performance during normal operation and safety during an attack. As an illustrative example, we have shown the effectiveness of the proposed design considering a multi-agent formation task involving 20 mobile robots. The simulation results testify that the safety filter operates effectively during false data injection and intelligent attacks.
☆ TJCCT: A Two-timescale Approach for UAV-assisted Mobile Edge Computing
Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) is emerging as a promising paradigm to provide aerial-terrestrial computing services in close proximity to mobile devices (MDs). However, meeting the demands of computation-intensive and delay-sensitive tasks for MDs poses several challenges, including the demand-supply contradiction between MDs and MEC servers, the demand-supply heterogeneity between MDs and MEC servers, the trajectory control requirements on energy efficiency and timeliness, and the different time-scale dynamics of the network. To address these issues, we first present a hierarchical architecture by incorporating terrestrial-aerial computing capabilities and leveraging UAV flexibility. Furthermore, we formulate a joint computing resource allocation, computation offloading, and trajectory control problem to maximize the system utility. Since the problem is a non-convex and NP-hard mixed integer nonlinear programming (MINLP), we propose a two-timescale joint computing resource allocation, computation offloading, and trajectory control (TJCCT) approach for solving the problem. In the short timescale, we propose a price-incentive model for on-demand computing resource allocation and a matching mechanism-based method for computation offloading. In the long timescale, we propose a convex optimization-based method for UAV trajectory control. Besides, we theoretically prove the stability, optimality, and polynomial complexity of TJCCT. Extended simulation results demonstrate that the proposed TJCCT outperforms the comparative algorithms in terms of the system utility, average processing rate, average completion delay, and average completion ratio.
☆ Scaling Learning based Policy Optimization for Temporal Tasks via Dropout
This paper introduces a model-based approach for training feedback controllers for an autonomous agent operating in a highly nonlinear environment. We desire the trained policy to ensure that the agent satisfies specific task objectives, expressed in discrete-time Signal Temporal Logic (DT-STL). One advantage for reformulation of a task via formal frameworks, like DT-STL, is that it permits quantitative satisfaction semantics. In other words, given a trajectory and a DT-STL formula, we can compute the robustness, which can be interpreted as an approximate signed distance between the trajectory and the set of trajectories satisfying the formula. We utilize feedback controllers, and we assume a feed forward neural network for learning these feedback controllers. We show how this learning problem is similar to training recurrent neural networks (RNNs), where the number of recurrent units is proportional to the temporal horizon of the agent's task objectives. This poses a challenge: RNNs are susceptible to vanishing and exploding gradients, and na\"{i}ve gradient descent-based strategies to solve long-horizon task objectives thus suffer from the same problems. To tackle this challenge, we introduce a novel gradient approximation algorithm based on the idea of dropout or gradient sampling. We show that, the existing smooth semantics for robustness are inefficient regarding gradient computation when the specification becomes complex. To address this challenge, we propose a new smooth semantics for DT-STL that under-approximates the robustness value and scales well for backpropagation over a complex specification. We show that our control synthesis methodology, can be quite helpful for stochastic gradient descent to converge with less numerical issues, enabling scalable backpropagation over long time horizons and trajectories over high dimensional state spaces.
☆ Semi-on-Demand Hybrid Transit Route Design with Shared Autonomous Mobility Services
This study examines the route design of a semi-on-demand hybrid route directional service in the public transit network, offering on-demand flexible route service in low-density areas and fixed route service in higher-density areas with Shared Autonomous Mobility Service (SAMS). The study develops analytically tractable cost expressions that capture access, waiting, and riding costs for users, and distance-based operating and time-based vehicle costs for operators. Two formulations are presented for strategic and tactical decisions in flexible route portion, fleet size, headway, and vehicle size optimization, enabling the determination of route types between fixed, hybrid, and flexible routes based on demand, cost, and operational parameters. The practical applications and benefits of semi-on-demand feeders are demonstrated with numerical examples and a large-scale case study in the Chicago metropolitan area. Findings reveal scenarios in which flexible route portions serving passengers located further away reduce total costs, particularly user costs. Lower operating costs in lower-demand areas favor more flexible routes, whereas higher demand densities favor more traditional line-based operations. On two studied lines, a current cost forecast favors smaller vehicles with flexible routes, but operating constraints and higher operating costs would favor bigger vehicles with hybrid routes. The study provides an analytical tool to design SAMS as directional services and transit feeders, and tractable continuous approximation formulations for future research in transit network design.
comment: 24 pages, 12 figures, the previous version presented at the 103rd Transportation Research Board Annual Meeting, Washington, D.C
☆ A Fairness-Oriented Reinforcement Learning Approach for the Operation and Control of Shared Micromobility Services
As Machine Learning systems become increasingly popular across diverse application domains, including those with direct human implications, the imperative of equity and algorithmic fairness has risen to prominence in the Artificial Intelligence community. On the other hand, in the context of Shared Micromobility Systems, the exploration of fairness-oriented approaches remains limited. Addressing this gap, we introduce a pioneering investigation into the balance between performance optimization and algorithmic fairness in the operation and control of Shared Micromobility Services. Our study leverages the Q-Learning algorithm in Reinforcement Learning, benefiting from its convergence guarantees to ensure the robustness of our proposed approach. Notably, our methodology stands out for its ability to achieve equitable outcomes, as measured by the Gini index, across different station categories--central, peripheral, and remote. Through strategic rebalancing of vehicle distribution, our approach aims to maximize operator performance while simultaneously upholding fairness principles for users. In addition to theoretical insights, we substantiate our findings with a case study or simulation based on synthetic data, validating the efficacy of our approach. This paper underscores the critical importance of fairness considerations in shaping control strategies for Shared Micromobility Services, offering a pragmatic framework for enhancing equity in urban transportation systems.
comment: 8 pages, 4 figures, submitted to the 63rd Conference on Decision and Control, Dec. 16-19, 2024, Milan, Italy
☆ Small Noise Analysis of Non-Parametric Closed-Loop Identification
We revisit the problem of non-parametric closed-loop identification in frequency domain; we give a brief survey of the literature and provide a small noise analysis of the direct, indirect, and joint input-output methods when two independent experiments with identical excitation are used. The analysis is asymptotic in the noise variance (i.e., as the standard deviation of the innovations $\sigma \to 0$), for a finite data record of length $N$. We highlight the relationship between the estimators accuracy and the loop shape via asymptotic variance expressions given in terms of the sensitivity function. The results are illustrated using a numerical simulation example.
☆ A Comparative Study of Artificial Potential Fields and Safety Filters
In this paper, we have demonstrated that the controllers designed by a classical motion planning tool, namely artificial potential fields (APFs), can be derived from a recently prevalent approach: control barrier function quadratic program (CBF-QP) safety filters. By integrating APF information into the CBF-QP framework, we establish a bridge between these two methodologies. Specifically, this is achieved by employing the attractive potential field as a control Lyapunov function (CLF) to guide the design of the nominal controller, and then the repulsive potential field serves as a reciprocal CBF (RCBF) to define a CBF-QP safety filter. Building on this integration, we extend the design of the CBF-QP safety filter to accommodate a more general class of dynamical models featuring a control-affine structure. This extension yields a special CBF-QP safety filter and a general APF solution suitable for control-affine dynamical models. Through a reach-avoid navigation example, we showcase the efficacy of the developed approaches.
☆ Distributed Robust Learning based Formation Control of Mobile Robots based on Bioinspired Neural Dynamics
This paper addresses the challenges of distributed formation control in multiple mobile robots, introducing a novel approach that enhances real-world practicability. We first introduce a distributed estimator using a variable structure and cascaded design technique, eliminating the need for derivative information to improve the real time performance. Then, a kinematic tracking control method is developed utilizing a bioinspired neural dynamic-based approach aimed at providing smooth control inputs and effectively resolving the speed jump issue. Furthermore, to address the challenges for robots operating with completely unknown dynamics and disturbances, a learning-based robust dynamic controller is developed. This controller provides real time parameter estimates while maintaining its robustness against disturbances. The overall stability of the proposed method is proved with rigorous mathematical analysis. At last, multiple comprehensive simulation studies have shown the advantages and effectiveness of the proposed method.
comment: This paper is accepted by IEEE Transactions on Intelligent Vehicles
☆ Causal Tracking of Distributions in Wasserstein Space: A Model Predictive Control Scheme
We consider the problem of optimal swarm tracking which can be formulated as a tracking problem for distributions in the Wasserstein metric. Optimal control solutions to this problem are non-causal and require knowing the time-trajectory of the distribution to be tracked in advance. We propose a scheme where these non-causal solutions can be used together with a predictive model for the reference to achieve causal tracking control of a priori-unknown references. We develop the resulting model-predictive control scheme in the simple case where the reference is predicted to be constant-in-time. A computational algorithm based on particle methods and discrete optimal mass transport is presented, and numerical simulations are provided for various classes of reference signals. The results demonstrate that the proposed control algorithm achieves reasonable performance even when using simple predictive models.
comment: 8 pages, 7 figures
☆ Improved Soft-k-Means Clustering Algorithm for Balancing Energy Consumption in Wireless Sensor Networks
Energy load balancing is an essential issue in designing wireless sensor networks (WSNs). Clustering techniques are utilized as energy-efficient methods to balance the network energy and prolong its lifetime. In this paper, we propose an improved soft-k-means (IS-k-means) clustering algorithm to balance the energy consumption of nodes in WSNs. First, we use the idea of ``clustering by fast search and find of density peaks'' (CFSFDP) and kernel density estimation (KDE) to improve the selection of the initial cluster centers of the soft k-means clustering algorithm. Then, we utilize the flexibility of the soft-k-means and reassign member nodes considering their membership probabilities at the boundary of clusters to balance the number of nodes per cluster. Furthermore, the concept of multi-cluster heads is employed to balance the energy consumption within clusters. {Extensive simulation results under different network scenarios demonstrate that for small-scale WSNs with single-hop transmission}, the proposed algorithm can postpone the first node death, the half of nodes death, and the last node death on average when compared to various clustering algorithms from the literature.
☆ Passivity-based Attack Identification and Mitigation with Event-triggered Observer Feedback and Switching Controller
This paper addresses the problem of output consensus in linear passive multi-agent systems under a False Data Injection (FDI) attack, considering the unavailability of complete state information. Our formulation relies on an event-based cryptographic authentication scheme for sensor integrity and considers FDI attacks at the actuator end, inspired by their practical nature and usages. For secure consensus, we propose (i) a passivity-based approach for detecting FDI attacks on the system and (ii) a Zeno-free event-triggered observer-based switching controller, which switches between the normal and the defense modes following an attack detection. We show that the closed-loop system achieves practical consensus under the controller's action in the defense mode. Simulation examples are provided to support the theoretical findings.
☆ Motion Planning for Identification of Linear Classifiers
A given region in 2-D Euclidean space is divided by a unknown linear classifier in to two sets each carrying a label. The objective of an agent with known dynamics traversing the region is to identify the true classifier while paying a control cost across its trajectory. We consider two scenarios: (i) the agent is able to measure the true label perfectly; (ii) the observed label is the true label multiplied by noise. We present the following: (i) the classifier identification problem formulated as a control problem; (ii) geometric interpretation of the control problem resulting in one step modified control problems; (iii) control algorithms that result in data sets which are used to identify the true classifier with accuracy; (iv) convergence of estimated classifier to the true classifier when the observed label is not corrupted by noise; (iv) numerical example demonstrating the utility of the control algorithms.
☆ On the role of network structure in learning to coordinate with bounded rationality IEEE Conference on Decision and Control
Many socioeconomic phenomena, such as technology adoption, collaborative problem-solving, and content engagement, involve a collection of agents coordinating to take a common action, aligning their decisions to maximize their individual goals. We consider a model for networked interactions where agents learn to coordinate their binary actions under a strict bound on their rationality. We first prove that our model is a potential game and that the optimal action profile is always to achieve perfect alignment at one of the two possible actions, regardless of the network structure. Using a stochastic learning algorithm known as Log Linear Learning, where agents have the same finite rationality parameter, we show that the probability of agents successfully agreeing on the correct decision is monotonically increasing in the number of network links. Therefore, more connectivity improves the accuracy of collective decision-making, as predicted by the phenomenon known as Wisdom of Crowds. Finally, we show that for a fixed number of links, a regular network maximizes the probability of success. We conclude that when using a network of irrational agents, promoting more homogeneous connectivity improves the accuracy of collective decision-making.
comment: Submitted to 2024 IEEE Conference on Decision and Control
☆ Safe and Stable Formation Control with Distributed Multi-Agents Using Adaptive Control and Control Barrier Functions
This manuscript considers the problem of ensuring stability and safety during formation control with distributed multi-agent systems in the presence of parametric uncertainty in the dynamics and limited communication. We propose an integrative approach that combines Control Barrier Functions, Adaptive Control, and connected graphs. A reference model is designed so as to ensure a safe and stable formation control strategy. This is combined with a provably correct adaptive control design that includes a use of a CBF-based safety filter that suitably generates safe reference commands, and employs error-based relaxation (EBR) of Nagumo's Invariance Theorem. Together, it is shown to lead to a guarantee of boundedness, formation control, and forward invariance. Numerical examples are provided to support the theoretical derivations.
☆ Real-Time Reconfiguration and Connectivity Maintenance for AUVs Network Under External Disturbances using Distributed Nonlinear Model Predictive Control
Advancements in underwater vehicle technology have significantly expanded the potential scope for deploying autonomous or remotely operated underwater vehicles in novel practical applications. However, the efficiency and maneuverability of these vehicles remain critical challenges, particularly in the dynamic aquatic environment. In this work, we propose a novel control scheme for creating multi-agent distributed formation control with limited communication between individual agents. In addition, the formation of the multi-agent can be reconfigured in real-time and the network connectivity can be maintained. The proposed use case for this scheme includes creating underwater mobile communication networks that can adapt to environmental or network conditions to maintain the quality of communication links for long-range exploration, seabed monitoring, or underwater infrastructure inspection. This work introduces a novel Distributed Nonlinear Model Predictive Control (DNMPC) strategy, integrating Control Lyapunov Functions (CLF) and Control Barrier Functions (CBF) with a relaxed decay rate, specifically tailored for 6-DOF underwater robotics. The effectiveness of our proposed DNMPC scheme was demonstrated through rigorous MATLAB simulations for trajectory tracking and formation reconfiguration in a dynamic environment. Our findings, supported by tests conducted using Software In The Loop (SITL) simulation, confirm the approach's applicability in real-time scenarios.
♻ ☆ On the impact of regularization in data-driven predictive control
Model predictive control (MPC) is a control strategy widely used in industrial applications. However, its implementation typically requires a mathematical model of the system being controlled, which can be a time-consuming and expensive task. Data-driven predictive control (DDPC) methods offer an alternative approach that does not require an explicit mathematical model, but instead optimize the control policy directly from data. In this paper, we study the impact of two different regularization penalties on the closed-loop performance of a recently introduced data-driven method called $\gamma$-DDPC. Moreover, we discuss the tuning of the related coefficients in different data and noise scenarios, to provide some guidelines for the end user.
comment: 7 pages, 3 figures
♻ ☆ Time-Varying Soft-Maximum Control Barrier Functions for Safety in an A Priori Unknown Environment
This paper presents a time-varying soft-maximum composite control barrier function (CBF) that can be used to ensure safety in an a priori unknown environment, where local perception information regarding the safe set is periodically obtained. We consider the scenario where the periodically obtained perception feedback can be used to construct a local CBF that models a local subset of the unknown safe set. Then, we use a novel smooth time-varying soft-maximum function to compose the N most recently obtained local CBFs into a single CBF. This composite CBF models an approximate union of the N most recently obtained local subsets of the safe set. Notably, this composite CBF can have arbitrary relative degree r. Next, this composite CBF is used as a rth-order CBF constraint in a real-time optimization to determine a control that minimizes a quadratic cost while guaranteeing that the state stays in a time-varying subset of the unknown safe set. We also present an application of the time-varying soft-maximum composite CBF method to a nonholonomic ground robot with nonnegligible inertia. In this application, we present a simple approach to generate the local CBFs from the periodically obtained perception data.
comment: 2024 American Control Conference (ACC)
♻ ☆ Hybrid Feedback Control for Global and Optimal Safe Navigation
We propose a hybrid feedback control strategy that safely steers a point-mass robot to a target location optimally from all initial conditions in the n-dimensional Euclidean space with a single spherical obstacle. The robot moves straight to the target when it has a clear line-of-sight to the target location. Otherwise, it engages in an optimal obstacle avoidance maneuver via the shortest path inside the cone enclosing the obstacle and having the robot's position as a vertex. The switching strategy that avoids the undesired equilibria, leading to global asymptotic stability (GAS) of the target location, relies on using two appropriately designed virtual destinations, ensuring control continuity and shortest path generation. Simulation results illustrating the effectiveness of the proposed approach are presented.
♻ ☆ Synthesis of Temporally-Robust Policies for Signal Temporal Logic Tasks using Reinforcement Learning ICRA 2024
This paper investigates the problem of designing control policies that satisfy high-level specifications described by signal temporal logic (STL) in unknown, stochastic environments. While many existing works concentrate on optimizing the spatial robustness of a system, our work takes a step further by also considering temporal robustness as a critical metric to quantify the tolerance of time uncertainty in STL. To this end, we formulate two relevant control objectives to enhance the temporal robustness of the synthesized policies. The first objective is to maximize the probability of being temporally robust for a given threshold. The second objective is to maximize the worst-case spatial robustness value within a bounded time shift. We use reinforcement learning to solve both control synthesis problems for unknown systems. Specifically, we approximate both control objectives in a way that enables us to apply the standard Q-learning algorithm. Theoretical bounds in terms of the approximations are also derived. We present case studies to demonstrate the feasibility of our approach.
comment: Accepted to ICRA 2024
♻ ☆ Topology Data Analysis-based Error Detection for Semantic Image Transmission with Incremental Knowledge-based HARQ
Semantic communication (SemCom) aims to achieve high fidelity information delivery under low communication consumption by only guaranteeing semantic accuracy. Nevertheless, semantic communication still suffers from unexpected channel volatility and thus developing a re-transmission mechanism (e.g., hybrid automatic repeat request [HARQ]) is indispensable. In that regard, instead of discarding previously transmitted information, the incremental knowledge-based HARQ (IK-HARQ) is deemed as a more effective mechanism that could sufficiently utilize the information semantics. However, considering the possible existence of semantic ambiguity in image transmission, a simple bit-level cyclic redundancy check (CRC) might compromise the performance of IK-HARQ. Therefore, it emerges a strong incentive to revolutionize the CRC mechanism, so as to reap the benefits of both SemCom and HARQ. In this paper, built on top of swin transformer-based joint source-channel coding (JSCC) and IK-HARQ, we propose a semantic image transmission framework SC-TDA-HARQ. In particular, different from the conventional CRC, we introduce a topological data analysis (TDA)-based error detection method, which capably digs out the inner topological and geometric information of images, so as to capture semantic information and determine the necessity for re-transmission. Extensive numerical results validate the effectiveness and efficiency of the proposed SC-TDA-HARQ framework, especially under the limited bandwidth condition, and manifest the superiority of TDA-based error detection method in image transmission.
♻ ☆ Hybrid path-lifting algorithm and Equivalence of Stability results for MRP-based control strategies
The modified Rodrigues parameters (MRP) consist of two numerically different triplets that, by switching between them, yield a minimal globally non-singular attitude description with advantageous properties. The MRP space results from the Alexandroff compactification of the three-dimensional Euclidean space and is a double cover of $\mathrm{SO(3)}$. By capitalizing on instrumental properties of the covering map, this paper proposes a novel hybrid dynamic path-lifting mechanism to unambiguously and robustly extract the MRP from the attitude space. This hybrid solution allows applying an MRP-based feedback controller to the attitude dynamics in the base space while preserving its asymptotic and exponential stability properties. Furthermore, by profiting from the distinct characteristics of the MRP, the resulting interconnection is impervious to the unwinding phenomenon. The design and validation of an MRP-based controller exemplify the application of the proposed algorithm alongside the novel results for equivalence of stability between spaces. The solution renders the attitude space tracking dynamics robustly globally exponentially stable, demonstrating the potential of this novel methodology.
♻ ☆ Performance-Guaranteed Solutions for Multi-Agent Optimal Coverage Problems using Submodularity, Curvature, and Greedy Algorithms
We consider a class of multi-agent optimal coverage problems in which the goal is to determine the optimal placement of a group of agents in a given mission space so that they maximize a coverage objective that represents a blend of individual and collaborative event detection capabilities. This class of problems is extremely challenging due to the non-convex nature of the mission space and of the coverage objective. With this motivation, greedy algorithms are often used as means of getting feasible coverage solutions efficiently. Even though such greedy solutions are suboptimal, the submodularity (diminishing returns) property of the coverage objective can be exploited to provide performance bound guarantees. Moreover, we show that improved performance bound guarantees (beyond the standard (1-1/e) performance bound) can be established using various curvature measures of the coverage problem. In particular, we provide a brief review of all existing popular applicable curvature measures, including a recent curvature measure that we proposed, and discuss their effectiveness and computational complexity, in the context of optimal coverage problems. We also propose novel computationally efficient techniques to estimate some curvature measures. Finally, we provide several numerical results to support our findings and propose several potential future research directions.
comment: Will be submitted to CDC 2024
Robotics 47
☆ Augmented Reality based Simulated Data (ARSim) with multi-view consistency for AV perception networks
Detecting a diverse range of objects under various driving scenarios is essential for the effectiveness of autonomous driving systems. However, the real-world data collected often lacks the necessary diversity presenting a long-tail distribution. Although synthetic data has been utilized to overcome this issue by generating virtual scenes, it faces hurdles such as a significant domain gap and the substantial efforts required from 3D artists to create realistic environments. To overcome these challenges, we present ARSim, a fully automated, comprehensive, modular framework designed to enhance real multi-view image data with 3D synthetic objects of interest. The proposed method integrates domain adaptation and randomization strategies to address covariate shift between real and simulated data by inferring essential domain attributes from real data and employing simulation-based randomization for other attributes. We construct a simplified virtual scene using real data and strategically place 3D synthetic assets within it. Illumination is achieved by estimating light distribution from multiple images capturing the surroundings of the vehicle. Camera parameters from real data are employed to render synthetic assets in each frame. The resulting augmented multi-view consistent dataset is used to train a multi-camera perception network for autonomous vehicles. Experimental results on various AV perception tasks demonstrate the superior performance of networks trained on the augmented dataset.
comment: 17 pages, 15 figures, 7 tables
☆ OceanPlan: Hierarchical Planning and Replanning for Natural Language AUV Piloting in Large-scale Unexplored Ocean Environments IROS 2024
We develop a hierarchical LLM-task-motion planning and replanning framework to efficiently ground an abstracted human command into tangible Autonomous Underwater Vehicle (AUV) control through enhanced representations of the world. We also incorporate a holistic replanner to provide real-world feedback with all planners for robust AUV operation. While there has been extensive research in bridging the gap between LLMs and robotic missions, they are unable to guarantee success of AUV applications in the vast and unknown ocean environment. To tackle specific challenges in marine robotics, we design a hierarchical planner to compose executable motion plans, which achieves planning efficiency and solution quality by decomposing long-horizon missions into sub-tasks. At the same time, real-time data stream is obtained by a replanner to address environmental uncertainties during plan execution. Experiments validate that our proposed framework delivers successful AUV performance of long-duration missions through natural language piloting.
comment: submitted to IROS 2024
Safe and Stable Teleoperation of Quadrotor UAVs under Haptic Shared Autonomy
We present a novel approach that aims to address both safety and stability of a haptic teleoperation system within a framework of Haptic Shared Autonomy (HSA). We use Control Barrier Functions (CBFs) to generate the control input that follows the user's input as closely as possible while guaranteeing safety. In the context of stability of the human-in-the-loop system, we limit the force feedback perceived by the user via a small $L_2$-gain, which is achieved by limiting the control and the force feedback via a differential constraint. Specifically, with the property of HSA, we propose two pathways to design the control and the force feedback: Sequential Control Force (SCF) and Joint Control Force (JCF). Both designs can achieve safety and stability but with different responses to the user's commands. We conducted experimental simulations to evaluate and investigate the properties of the designed methods. We also tested the proposed method on a physical quadrotor UAV and a haptic interface.
☆ Gesture-Controlled Aerial Robot Formation for Human-Swarm Interaction in Safety Monitoring Applications
This paper presents a formation control approach for contactless gesture-based Human-Swarm Interaction (HSI) between a team of multi-rotor Unmanned Aerial Vehicles (UAVs) and a human worker. The approach is intended for monitoring the safety of human workers, especially those working at heights. In the proposed dynamic formation scheme, one UAV acts as the leader of the formation and is equipped with sensors for human worker detection and gesture recognition. The follower UAVs maintain a predetermined formation relative to the worker's position, thereby providing additional perspectives of the monitored scene. Hand gestures allow the human worker to specify movements and action commands for the UAV team and initiate other mission-related commands without the need for an additional communication channel or specific markers. Together with a novel unified human detection and tracking algorithm, human pose estimation approach and gesture detection pipeline, the proposed approach forms a first instance of an HSI system incorporating all these modules onboard real-world UAVs. Simulations and field experiments with three UAVs and a human worker in a mock-up scenario showcase the effectiveness and responsiveness of the proposed approach.
comment: 8 pages, 9 figures
☆ Introduction to Human-Robot Interaction: A Multi-Perspective Introductory Course
In this paper I describe the design of an introductory course in Human-Robot Interaction. This project-driven course is designed to introduce undergraduate and graduate engineering students, especially those enrolled in Computer Science, Mechanical Engineering, and Robotics degree programs, to key theories and methods used in the field of Human-Robot Interaction that they would otherwise be unlikely to see in those degree programs. To achieve this aim, the course takes students all the way from stakeholder analysis to empirical evaluation, covering and integrating key Qualitative, Design, Computational, and Quantitative methods along the way. I detail the goals, audience, and format of the course, and provide a detailed walkthrough of the course syllabus.
comment: Presented at the Designing an Intro to HRI Course Workshop at HRI 2024 (arXiv:2403.05588)
☆ HortiBot: An Adaptive Multi-Arm System for Robotic Horticulture of Sweet Peppers IROS
Horticultural tasks such as pruning and selective harvesting are labor intensive and horticultural staff are hard to find. Automating these tasks is challenging due to the semi-structured greenhouse workspaces, changing environmental conditions such as lighting, dense plant growth with many occlusions, and the need for gentle manipulation of non-rigid plant organs. In this work, we present the three-armed system HortiBot, with two arms for manipulation and a third arm as an articulated head for active perception using stereo cameras. Its perception system detects not only peppers, but also peduncles and stems in real time, and performs online data association to build a world model of pepper plants. Collision-aware online trajectory generation allows all three arms to safely track their respective targets for observation, grasping, and cutting. We integrated perception and manipulation to perform selective harvesting of peppers and evaluated the system in lab experiments. Using active perception coupled with end-effector force torque sensing for compliant manipulation, HortiBot achieves high success rates.
comment: Submitted to International Conference on Intelligent Robots and Systems (IROS) 2024. C. Lenz and R. Menon contributed equally
☆ Guided Decoding for Robot Motion Generation and Adaption
We address motion generation for high-DoF robot arms in complex settings with obstacles, via points, etc. A significant advancement in this domain is achieved by integrating Learning from Demonstration (LfD) into the motion generation process. This integration facilitates rapid adaptation to new tasks and optimizes the utilization of accumulated expertise by allowing robots to learn and generalize from demonstrated trajectories. We train a transformer architecture on a large dataset of simulated trajectories. This architecture, based on a conditional variational autoencoder transformer, learns essential motion generation skills and adapts these to meet auxiliary tasks and constraints. Our auto-regressive approach enables real-time integration of feedback from the physical system, enhancing the adaptability and efficiency of motion generation. We show that our model can generate motion from initial and target points, but also that it can adapt trajectories in navigating complex tasks, including obstacle avoidance, via points, and meeting velocity and acceleration constraints, across platforms.
comment: 7 pages
☆ TriHelper: Zero-Shot Object Navigation with Dynamic Assistance
Navigating toward specific objects in unknown environments without additional training, known as Zero-Shot object navigation, poses a significant challenge in the field of robotics, which demands high levels of auxiliary information and strategic planning. Traditional works have focused on holistic solutions, overlooking the specific challenges agents encounter during navigation such as collision, low exploration efficiency, and misidentification of targets. To address these challenges, our work proposes TriHelper, a novel framework designed to assist agents dynamically through three primary navigation challenges: collision, exploration, and detection. Specifically, our framework consists of three innovative components: (i) Collision Helper, (ii) Exploration Helper, and (iii) Detection Helper. These components work collaboratively to solve these challenges throughout the navigation process. Experiments on the Habitat-Matterport 3D (HM3D) and Gibson datasets demonstrate that TriHelper significantly outperforms all existing baseline methods in Zero-Shot object navigation, showcasing superior success rates and exploration efficiency. Our ablation studies further underscore the effectiveness of each helper in addressing their respective challenges, notably enhancing the agent's navigation capabilities. By proposing TriHelper, we offer a fresh perspective on advancing the object navigation task, paving the way for future research in the domain of Embodied AI and visual-based navigation.
comment: 8 pages, 5 figures
☆ DITTO: Demonstration Imitation by Trajectory Transformation IROS 2024
Teaching robots new skills quickly and conveniently is crucial for the broader adoption of robotic systems. In this work, we address the problem of one-shot imitation from a single human demonstration, given by an RGB-D video recording through a two-stage process. In the first stage which is offline, we extract the trajectory of the demonstration. This entails segmenting manipulated objects and determining their relative motion in relation to secondary objects such as containers. Subsequently, in the live online trajectory generation stage, we first \mbox{re-detect} all objects, then we warp the demonstration trajectory to the current scene, and finally, we trace the trajectory with the robot. To complete these steps, our method makes leverages several ancillary models, including those for segmentation, relative object pose estimation, and grasp prediction. We systematically evaluate different combinations of correspondence and re-detection methods to validate our design decision across a diverse range of tasks. Specifically, we collect demonstrations of ten different tasks including pick-and-place tasks as well as articulated object manipulation. Finally, we perform extensive evaluations on a real robot system to demonstrate the effectiveness and utility of our approach in real-world scenarios. We make the code publicly available at http://ditto.cs.uni-freiburg.de.
comment: 8 pages, 4 figures, 3 tables, submitted to IROS 2024
☆ CRPlace: Camera-Radar Fusion with BEV Representation for Place Recognition
The integration of complementary characteristics from camera and radar data has emerged as an effective approach in 3D object detection. However, such fusion-based methods remain unexplored for place recognition, an equally important task for autonomous systems. Given that place recognition relies on the similarity between a query scene and the corresponding candidate scene, the stationary background of a scene is expected to play a crucial role in the task. As such, current well-designed camera-radar fusion methods for 3D object detection can hardly take effect in place recognition because they mainly focus on dynamic foreground objects. In this paper, a background-attentive camera-radar fusion-based method, named CRPlace, is proposed to generate background-attentive global descriptors from multi-view images and radar point clouds for accurate place recognition. To extract stationary background features effectively, we design an adaptive module that generates the background-attentive mask by utilizing the camera BEV feature and radar dynamic points. With the guidance of a background mask, we devise a bidirectional cross-attention-based spatial fusion strategy to facilitate comprehensive spatial interaction between the background information of the camera BEV feature and the radar BEV feature. As the first camera-radar fusion-based place recognition network, CRPlace has been evaluated thoroughly on the nuScenes dataset. The results show that our algorithm outperforms a variety of baseline methods across a comprehensive set of metrics (recall@1 reaches 91.2%).
☆ AV-Occupant Perceived Risk Model for Cut-In Scenarios with Empirical Evaluation
Advancements in autonomous vehicle (AV) technologies necessitate precise estimation of perceived risk to enhance user comfort, acceptance and trust. This paper introduces a novel AV-Occupant Risk (AVOR) model designed for perceived risk estimation during AV cut-in scenarios. An empirical study is conducted with 18 participants with realistic cut-in scenarios. Two factors were investigated: scenario risk and scene population. 76% of subjective risk responses indicate an increase in perceived risk at cut-in initiation. The existing perceived risk model did not capture this critical phenomenon. Our AVOR model demonstrated a significant improvement in estimating perceived risk during the early stages of cut-ins, especially for the high-risk scenario, enhancing modelling accuracy by up to 54%. The concept of the AVOR model can quantify perceived risk in other diverse driving contexts characterized by dynamic uncertainties, enhancing the reliability and human-centred focus of AV systems.
☆ Infrastructure-Assisted Collaborative Perception in Automated Valet Parking: A Safety Perspective
Environmental perception in Automated Valet Parking (AVP) has been a challenging task due to severe occlusions in parking garages. Although Collaborative Perception (CP) can be applied to broaden the field of view of connected vehicles, the limited bandwidth of vehicular communications restricts its application. In this work, we propose a BEV feature-based CP network architecture for infrastructure-assisted AVP systems. The model takes the roadside camera and LiDAR as optional inputs and adaptively fuses them with onboard sensors in a unified BEV representation. Autoencoder and downsampling are applied for channel-wise and spatial-wise dimension reduction, while sparsification and quantization further compress the feature map with little loss in data precision. Combining these techniques, the size of a BEV feature map is effectively compressed to fit in the feasible data rate of the NR-V2X network. With the synthetic AVP dataset, we observe that CP can effectively increase perception performance, especially for pedestrians. Moreover, the advantage of infrastructure-assisted CP is demonstrated in two typical safety-critical scenarios in the AVP setting, increasing the maximum safe cruising speed by up to 3m/s in both scenarios.
comment: 7 pages, 7 figures, 4 tables, accepted by IEEE VTC2024-Spring
☆ RHINO-VR Experience: Teaching Mobile Robotics Concepts in an Interactive Museum Exhibit
In 1997, the very first tour guide robot RHINO was deployed in a museum in Germany. With the ability to navigate autonomously through the environment, the robot gave tours to over 2,000 visitors. Today, RHINO itself has become an exhibit and is no longer operational. In this paper, we present RHINO-VR, an interactive museum exhibit using virtual reality (VR) that allows museum visitors to experience the historical robot RHINO in operation in a virtual museum. RHINO-VR, unlike static exhibits, enables users to familiarize themselves with basic mobile robotics concepts without the fear of damaging the exhibit. In the virtual environment, the user is able to interact with RHINO in VR by pointing to a location to which the robot should navigate and observing the corresponding actions of the robot. To include other visitors who cannot use the VR, we provide an external observation view to make RHINO visible to them. We evaluated our system by measuring the frame rate of the VR simulation, comparing the generated virtual 3D models with the originals, and conducting a user study. The user-study showed that RHINO-VR improved the visitors' understanding of the robot's functionality and that they would recommend experiencing the VR exhibit to others.
comment: Submitted to IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN)
☆ ALPINE: a climbing robot for operations in mountain environments
Mountain slopes are perfect examples of harsh environments in which humans are required to perform difficult and dangerous operations such as removing unstable boulders, dangerous vegetation or deploying safety nets. A good replacement for human intervention can be offered by climbing robots. The different solutions existing in the literature are not up to the task for the difficulty of the requirements (navigation, heavy payloads, flexibility in the execution of the tasks). In this paper, we propose a robotic platform that can fill this gap. Our solution is based on a robot that hangs on ropes, and uses a retractable leg to jump away from the mountain walls. Our package of mechanical solutions, along with the algorithms developed for motion planning and control, delivers swift navigation on irregular and steep slopes, the possibility to overcome or travel around significant natural barriers, and the ability to carry heavy payloads and execute complex tasks. In the paper, we give a full account of our main design and algorithmic choices and show the feasibility of the solution through a large number of physically simulated scenarios.
☆ Collision Avoidance Safety Filter for an Autonomous E-Scooter using Ultrasonic Sensors
In this paper, we propose a collision avoidance safety filter for autonomous electric scooters to enable safe operation of such vehicles in pedestrian areas. In particular, we employ multiple low-cost ultrasonic sensors to detect a wide range of possible obstacles in front of the e-scooter. Based on possibly faulty distance measurements, we design a filter to mitigate measurement noise and missing values as well as a gain-scheduled controller to limit the velocity commanded to the e-scooter when required due to imminent collisions. The proposed controller structure is able to prevent collisions with unknown obstacles by deploying a reduced safe velocity ensuring a sufficiently large safety distance. The collision avoidance approach is designed such that it may be easily deployed in similar applications of general micromobility vehicles. The effectiveness of our proposed safety filter is demonstrated in real-world experiments.
☆ Set-membership target search and tracking within an unknown cluttered area using cooperating UAVs equipped with vision systems
This paper addresses the problem of target search and tracking using a fleet of cooperating UAVs evolving in some unknown region of interest containing an a priori unknown number of moving ground targets. Each drone is equipped with an embedded Computer Vision System (CVS), providing an image with labeled pixels and a depth map of the observed part of its environment. Moreover, a box containing the corresponding pixels in the image frame is available when a UAV identifies a target. Hypotheses regarding information provided by the pixel classification, depth map construction, and target identification algorithms are proposed to allow its exploitation by set-membership approaches. A set-membership target location estimator is developed using the information provided by the CVS. Each UAV evaluates sets guaranteed to contain the location of the identified targets and a set possibly containing the locations of targets still to be identified. Then, each UAV uses these sets to search and track targets cooperatively.
☆ PseudoTouch: Efficiently Imaging the Surface Feel of Objects for Robotic Manipulation IROS2024
Humans seemingly incorporate potential touch signals in their perception. Our goal is to equip robots with a similar capability, which we term \ourmodel. \ourmodel aims to predict the expected touch signal based on a visual patch representing the touched area. We frame this problem as the task of learning a low-dimensional visual-tactile embedding, wherein we encode a depth patch from which we decode the tactile signal. To accomplish this task, we employ ReSkin, an inexpensive and replaceable magnetic-based tactile sensor. Using ReSkin, we collect and train PseudoTouch on a dataset comprising aligned tactile and visual data pairs obtained through random touching of eight basic geometric shapes. We demonstrate the efficacy of PseudoTouch through its application to two downstream tasks: object recognition and grasp stability prediction. In the object recognition task, we evaluate the learned embedding's performance on a set of five basic geometric shapes and five household objects. Using PseudoTouch, we achieve an object recognition accuracy 84% after just ten touches, surpassing a proprioception baseline. For the grasp stability task, we use ACRONYM labels to train and evaluate a grasp success predictor using PseudoTouch's predictions derived from virtual depth information. Our approach yields an impressive 32% absolute improvement in accuracy compared to the baseline relying on partial point cloud data. We make the data, code, and trained models publicly available at http://pseudotouch.cs.uni-freiburg.de.
comment: 8 pages, 7 figures, 2 tables, submitted to IROS2024
☆ Learning from Visual Demonstrations through Differentiable Nonlinear MPC for Personalized Autonomous Driving
Human-like autonomous driving controllers have the potential to enhance passenger perception of autonomous vehicles. This paper proposes DriViDOC: a model for Driving from Vision through Differentiable Optimal Control, and its application to learn personalized autonomous driving controllers from human demonstrations. DriViDOC combines the automatic inference of relevant features from camera frames with the properties of nonlinear model predictive control (NMPC), such as constraint satisfaction. Our approach leverages the differentiability of parametric NMPC, allowing for end-to-end learning of the driving model from images to control. The model is trained on an offline dataset comprising various driving styles collected on a motion-base driving simulator. During online testing, the model demonstrates successful imitation of different driving styles, and the interpreted NMPC parameters provide insights into the achievement of specific driving behaviors. Our experimental results show that DriViDOC outperforms other methods involving NMPC and neural networks, exhibiting an average improvement of 20% in imitation scores.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. Accompanying video available at: https://youtu.be/WxWPuAtJ08E
☆ Subequivariant Reinforcement Learning Framework for Coordinated Motion Control
Effective coordination is crucial for motion control with reinforcement learning, especially as the complexity of agents and their motions increases. However, many existing methods struggle to account for the intricate dependencies between joints. We introduce CoordiGraph, a novel architecture that leverages subequivariant principles from physics to enhance coordination of motion control with reinforcement learning. This method embeds the principles of equivariance as inherent patterns in the learning process under gravity influence, which aids in modeling the nuanced relationships between joints vital for motion control. Through extensive experimentation with sophisticated agents in diverse environments, we highlight the merits of our approach. Compared to current leading methods, CoordiGraph notably enhances generalization and sample efficiency.
comment: 7 pages, 7 figures, 2024 IEEE International Conference on Robotics and Automation
☆ Automated Feature Selection for Inverse Reinforcement Learning
Inverse reinforcement learning (IRL) is an imitation learning approach to learning reward functions from expert demonstrations. Its use avoids the difficult and tedious procedure of manual reward specification while retaining the generalization power of reinforcement learning. In IRL, the reward is usually represented as a linear combination of features. In continuous state spaces, the state variables alone are not sufficiently rich to be used as features, but which features are good is not known in general. To address this issue, we propose a method that employs polynomial basis functions to form a candidate set of features, which are shown to allow the matching of statistical moments of state distributions. Feature selection is then performed for the candidates by leveraging the correlation between trajectory probabilities and feature expectations. We demonstrate the approach's effectiveness by recovering reward functions that capture expert policies across non-linear control tasks of increasing complexity. Code, data, and videos are available at https://sites.google.com/view/feature4irl.
comment: 7 pages, 4 figures
☆ A Twin Delayed Deep Deterministic Policy Gradient Algorithm for Autonomous Ground Vehicle Navigation via Digital Twin Perception Awareness
Autonomous ground vehicle (UGV) navigation has the potential to revolutionize the transportation system by increasing accessibility to disabled people, ensure safety and convenience of use. However, UGV requires extensive and efficient testing and evaluation to ensure its acceptance for public use. This testing are mostly done in a simulator which result to sim2real transfer gap. In this paper, we propose a digital twin perception awareness approach for the control of robot navigation without prior creation of the virtual environment (VT) environment state. To achieve this, we develop a twin delayed deep deterministic policy gradient (TD3) algorithm that ensures collision avoidance and goal-based path planning. We demonstrate the performance of our approach on different environment dynamics. We show that our approach is capable of efficiently avoiding collision with obstacles and navigating to its desired destination, while at the same time safely avoids obstacles using the information received from the LIDAR sensor mounted on the robot. Our approach bridges the gap between sim-to-real transfer and contributes to the adoption of UGVs in real world. We validate our approach in simulation and a real-world application in an office space.
comment: 8 pages, 7 figures
☆ Rethinking 6-Dof Grasp Detection: A Flexible Framework for High-Quality Grasping
Robotic grasping is a primitive skill for complex tasks and is fundamental to intelligence. For general 6-Dof grasping, most previous methods directly extract scene-level semantic or geometric information, while few of them consider the suitability for various downstream applications, such as target-oriented grasping. Addressing this issue, we rethink 6-Dof grasp detection from a grasp-centric view and propose a versatile grasp framework capable of handling both scene-level and target-oriented grasping. Our framework, FlexLoG, is composed of a Flexible Guidance Module and a Local Grasp Model. Specifically, the Flexible Guidance Module is compatible with both global (e.g., grasp heatmap) and local (e.g., visual grounding) guidance, enabling the generation of high-quality grasps across various tasks. The Local Grasp Model focuses on object-agnostic regional points and predicts grasps locally and intently. Experiment results reveal that our framework achieves over 18% and 23% improvement on unseen splits of the GraspNet-1Billion Dataset. Furthermore, real-world robotic tests in three distinct settings yield a 95% success rate.
comment: 8 pages, 8 figures
☆ Linear Quadratic Guidance Law for Joint Motion Planning of a Pursuer-Turret Assembly
This paper presents joint motion planning of a vehicle with an attached rotating turret. The turret has a limited range as well as the field of view. The objective is capture a maneuvering target such that at the terminal time it is withing the field-of-view and range limits. Catering to it, we present a minimum effort guidance law that commensurate for the turn rate abilities of the vehicle and the turret. The guidance law is obtained using linearization about the collision triangle and admits an analytical solution. Simulation results are presented to exemplify the cooperation between the turret and the vehicle.
☆ Boundary-Aware Value Function Generation for Safe Stochastic Motion Planning
Navigation safety is critical for many autonomous systems such as self-driving vehicles in an urban environment. It requires an explicit consideration of boundary constraints that describe the borders of any infeasible, non-navigable, or unsafe regions. We propose a principled boundary-aware safe stochastic planning framework with promising results. Our method generates a value function that can strictly distinguish the state values between free (safe) and non-navigable (boundary) spaces in the continuous state, naturally leading to a safe boundary-aware policy. At the core of our solution lies a seamless integration of finite elements and kernel-based functions, where the finite elements allow us to characterize safety-critical states' borders accurately, and the kernel-based function speeds up computation for the non-safety-critical states. The proposed method was evaluated through extensive simulations and demonstrated safe navigation behaviors in mobile navigation tasks. Additionally, we demonstrate that our approach can maneuver safely and efficiently in cluttered real-world environments using a ground vehicle with strong external disturbances, such as navigating on a slippery floor and against external human intervention.
comment: Accepted by International Journal of Robotics Research
☆ SRLM: Human-in-Loop Interactive Social Robot Navigation with Large Language Model and Deep Reinforcement Learning
An interactive social robotic assistant must provide services in complex and crowded spaces while adapting its behavior based on real-time human language commands or feedback. In this paper, we propose a novel hybrid approach called Social Robot Planner (SRLM), which integrates Large Language Models (LLM) and Deep Reinforcement Learning (DRL) to navigate through human-filled public spaces and provide multiple social services. SRLM infers global planning from human-in-loop commands in real-time, and encodes social information into a LLM-based large navigation model (LNM) for low-level motion execution. Moreover, a DRL-based planner is designed to maintain benchmarking performance, which is blended with LNM by a large feedback model (LFM) to address the instability of current text and LLM-driven LNM. Finally, SRLM demonstrates outstanding performance in extensive experiments. More details about this work are available at: https://sites.google.com/view/navi-srlm
☆ CoNVOI: Context-aware Navigation using Vision Language Models in Outdoor and Indoor Environments
We present ConVOI, a novel method for autonomous robot navigation in real-world indoor and outdoor environments using Vision Language Models (VLMs). We employ VLMs in two ways: first, we leverage their zero-shot image classification capability to identify the context or scenario (e.g., indoor corridor, outdoor terrain, crosswalk, etc) of the robot's surroundings, and formulate context-based navigation behaviors as simple text prompts (e.g. ``stay on the pavement"). Second, we utilize their state-of-the-art semantic understanding and logical reasoning capabilities to compute a suitable trajectory given the identified context. To this end, we propose a novel multi-modal visual marking approach to annotate the obstacle-free regions in the RGB image used as input to the VLM with numbers, by correlating it with a local occupancy map of the environment. The marked numbers ground image locations in the real-world, direct the VLM's attention solely to navigable locations, and elucidate the spatial relationships between them and terrains depicted in the image to the VLM. Next, we query the VLM to select numbers on the marked image that satisfy the context-based behavior text prompt, and construct a reference path using the selected numbers. Finally, we propose a method to extrapolate the reference trajectory when the robot's environmental context has not changed to prevent unnecessary VLM queries. We use the reference trajectory to guide a motion planner, and demonstrate that it leads to human-like behaviors (e.g. not cutting through a group of people, using crosswalks, etc.) in various real-world indoor and outdoor scenarios.
comment: 9 pages, 4 figures
☆ Global Games with Negative Feedback for Autonomous Colony Maintenance using Robot Teams
In this article we address the colony maintenance problem, where a team of robots are tasked with continuously maintaining the energy supply of an autonomous colony. We model this as a global game, where robots measure the energy level of a central nest to determine whether or not to forage for energy sources. We design a mechanism that avoids the trivial equilibrium where all robots always forage. Furthermore, we demonstrate that when the game is played iteratively a negative feedback term stabilizes the number of foraging robots at a non-trivial Nash equilibrium. We compare our approach qualitatively to existing global games, where a positive positive feedback term admits threshold-based decision making, and encourages many robots to forage simultaneously. We discuss how positive feedback can lead to a cascading failure in the presence of a human who recruits robots for external tasks, and we demonstrate the performance of our approach in simulation.
comment: 6 pages, 5 figures
☆ Autonomous Driving With Perception Uncertainties: Deep-Ensemble Based Adaptive Cruise Control
Autonomous driving depends on perception systems to understand the environment and to inform downstream decision-making. While advanced perception systems utilizing black-box Deep Neural Networks (DNNs) demonstrate human-like comprehension, their unpredictable behavior and lack of interpretability may hinder their deployment in safety critical scenarios. In this paper, we develop an Ensemble of DNN regressors (Deep Ensemble) that generates predictions with quantification of prediction uncertainties. In the scenario of Adaptive Cruise Control (ACC), we employ the Deep Ensemble to estimate distance headway to the lead vehicle from RGB images and enable the downstream controller to account for the estimation uncertainty. We develop an adaptive cruise controller that utilizes Stochastic Model Predictive Control (MPC) with chance constraints to provide a probabilistic safety guarantee. We evaluate our ACC algorithm using a high-fidelity traffic simulator and a real-world traffic dataset and demonstrate the ability of the proposed approach to effect speed tracking and car following while maintaining a safe distance headway. The out-of-distribution scenarios are also examined.
☆ Music to Dance as Language Translation using Sequence Models
Synthesising appropriate choreographies from music remains an open problem. We introduce MDLT, a novel approach that frames the choreography generation problem as a translation task. Our method leverages an existing data set to learn to translate sequences of audio into corresponding dance poses. We present two variants of MDLT: one utilising the Transformer architecture and the other employing the Mamba architecture. We train our method on AIST++ and PhantomDance data sets to teach a robotic arm to dance, but our method can be applied to a full humanoid robot. Evaluation metrics, including Average Joint Error and Frechet Inception Distance, consistently demonstrate that, when given a piece of music, MDLT excels at producing realistic and high-quality choreography. The code can be found at github.com/meowatthemoon/MDLT.
♻ ☆ Gaussian-SLAM: Photo-realistic Dense SLAM with Gaussian Splatting
We present a dense simultaneous localization and mapping (SLAM) method that uses 3D Gaussians as a scene representation. Our approach enables interactive-time reconstruction and photo-realistic rendering from real-world single-camera RGBD videos. To this end, we propose a novel effective strategy for seeding new Gaussians for newly explored areas and their effective online optimization that is independent of the scene size and thus scalable to larger scenes. This is achieved by organizing the scene into sub-maps which are independently optimized and do not need to be kept in memory. We further accomplish frame-to-model camera tracking by minimizing photometric and geometric losses between the input and rendered frames. The Gaussian representation allows for high-quality photo-realistic real-time rendering of real-world scenes. Evaluation on synthetic and real-world datasets demonstrates competitive or superior performance in mapping, tracking, and rendering compared to existing neural dense SLAM methods.
♻ ☆ A Convex Formulation of Frictional Contact for the Material Point Method and Rigid Bodies
In this paper, we introduce a novel convex formulation that seamlessly integrates the Material Point Method (MPM) with articulated rigid body dynamics in frictional contact scenarios. We extend the linear corotational hyperelastic model into the realm of elastoplasticity and include an efficient return mapping algorithm. This approach is particularly effective for MPM simulations involving significant deformation and topology changes, while preserving the convexity of the optimization problem. Our method ensures global convergence, enabling the use of large simulation time steps without compromising robustness. We have validated our approach through rigorous testing and performance evaluations, highlighting its superior capabilities in managing complex simulations relevant to robotics. Compared to previous MPM based robotic simulators, our method significantly improves the stability of contact resolution -- a critical factor in robot manipulation tasks. We make our method available in the open-source robotics toolkit, Drake.
comment: The supplemental video is available at https://youtu.be/5jrQtF5D0DA
♻ ☆ Learning High-level Semantic-Relational Concepts for SLAM
Recent works on SLAM extend their pose graphs with higher-level semantic concepts like Rooms exploiting relationships between them, to provide, not only a richer representation of the situation/environment but also to improve the accuracy of its estimation. Concretely, our previous work, Situational Graphs (S-Graphs+), a pioneer in jointly leveraging semantic relationships in the factor optimization process, relies on semantic entities such as Planes and Rooms, whose relationship is mathematically defined. Nevertheless, there is no unique approach to finding all the hidden patterns in lower-level factor-graphs that correspond to high-level concepts of different natures. It is currently tackled with ad-hoc algorithms, which limits its graph expressiveness. To overcome this limitation, in this work, we propose an algorithm based on Graph Neural Networks for learning high-level semantic-relational concepts that can be inferred from the low-level factor graph. Given a set of mapped Planes our algorithm is capable of inferring Room entities relating to the Planes. Additionally, to demonstrate the versatility of our method, our algorithm can infer an additional semantic-relational concept, i.e. Wall, and its relationship with its Planes. We validate our method in both simulated and real datasets demonstrating improved performance over two baseline approaches. Furthermore, we integrate our method into the S-Graphs+ algorithm providing improved pose and map accuracy compared to the baseline while further enhancing the scene representation.
♻ ☆ LaMI: Large Language Models for Multi-Modal Human-Robot Interaction
This paper presents an innovative large language model (LLM)-based robotic system for enhancing multi-modal human-robot interaction (HRI). Traditional HRI systems relied on complex designs for intent estimation, reasoning, and behavior generation, which were resource-intensive. In contrast, our system empowers researchers and practitioners to regulate robot behavior through three key aspects: providing high-level linguistic guidance, creating "atomic actions" and expressions the robot can use, and offering a set of examples. Implemented on a physical robot, it demonstrates proficiency in adapting to multi-modal inputs and determining the appropriate manner of action to assist humans with its arms, following researchers' defined guidelines. Simultaneously, it coordinates the robot's lid, neck, and ear movements with speech output to produce dynamic, multi-modal expressions. This showcases the system's potential to revolutionize HRI by shifting from conventional, manual state-and-flow design methods to an intuitive, guidance-based, and example-driven approach. Supplementary material can be found at https://hri-eu.github.io/Lami/
comment: 10 pages, 6 figures
♻ ☆ Bi-KVIL: Keypoints-based Visual Imitation Learning of Bimanual Manipulation Tasks
Visual imitation learning has achieved impressive progress in learning unimanual manipulation tasks from a small set of visual observations, thanks to the latest advances in computer vision. However, learning bimanual coordination strategies and complex object relations from bimanual visual demonstrations, as well as generalizing them to categorical objects in novel cluttered scenes remain unsolved challenges. In this paper, we extend our previous work on keypoints-based visual imitation learning (\mbox{K-VIL})~\cite{gao_kvil_2023} to bimanual manipulation tasks. The proposed Bi-KVIL jointly extracts so-called \emph{Hybrid Master-Slave Relationships} (HMSR) among objects and hands, bimanual coordination strategies, and sub-symbolic task representations. Our bimanual task representation is object-centric, embodiment-independent, and viewpoint-invariant, thus generalizing well to categorical objects in novel scenes. We evaluate our approach in various real-world applications, showcasing its ability to learn fine-grained bimanual manipulation tasks from a small number of human demonstration videos. Videos and source code are available at https://sites.google.com/view/bi-kvil.
♻ ☆ Event-based Simultaneous Localization and Mapping: A Comprehensive Survey
In recent decades, visual simultaneous localization and mapping (vSLAM) has gained significant interest in both academia and industry. It estimates camera motion and reconstructs the environment concurrently using visual sensors on a moving robot. However, conventional cameras are limited by hardware, including motion blur and low dynamic range, which can negatively impact performance in challenging scenarios like high-speed motion and high dynamic range illumination. Recent studies have demonstrated that event cameras, a new type of bio-inspired visual sensor, offer advantages such as high temporal resolution, dynamic range, low power consumption, and low latency. This paper presents a timely and comprehensive review of event-based vSLAM algorithms that exploit the benefits of asynchronous and irregular event streams for localization and mapping tasks. The review covers the working principle of event cameras and various event representations for preprocessing event data. It also categorizes event-based vSLAM methods into four main categories: feature-based, direct, motion-compensation, and deep learning methods, with detailed discussions and practical guidance for each approach. Furthermore, the paper evaluates the state-of-the-art methods on various benchmarks, highlighting current challenges and future opportunities in this emerging research area. A public repository will be maintained to keep track of the rapid developments in this field at {\url{https://github.com/kun150kun/ESLAM-survey}}.
♻ ☆ Robust Direct Data-Driven Control for Probabilistic Systems
We propose a data-driven control method for systems with aleatoric uncertainty, for example, robot fleets with variations between agents. Our method leverages shared trajectory data to increase the robustness of the designed controller and thus facilitate transfer to new variations without the need for prior parameter and uncertainty estimations. In contrast to existing work on experience transfer for performance, our approach focuses on robustness and uses data collected from multiple realizations to guarantee generalization to unseen ones. Our method is based on scenario optimization combined with recent formulations for direct data-driven control. We derive lower bounds on the amount of data required to achieve quadratic stability for probabilistic systems with aleatoric uncertainty and demonstrate the benefits of our data-driven method through a numerical example. We find that the learned controllers generalize well to high variations in the dynamics even when based on only a few short open-loop trajectories. Robust experience transfer enables the design of safe and robust controllers that work out of the box without any additional learning during deployment.
♻ ☆ A Wind-Aware Path Planning Method for UAV-Asisted Bridge Inspection
In response to the gap in considering wind conditions in the bridge inspection using unmanned aerial vehicle (UAV) , this paper proposes a path planning method for UAVs that takes into account the influence of wind, based on the simulated annealing algorithm. The algorithm considers the wind factors, including the influence of different wind speeds and directions at the same time on the path planning of the UAV. Firstly, An environment model is constructed specifically for UAV bridge inspection, taking into account the various objective functions and constraint conditions of UAVs. A more sophisticated and precise mathematical model is then developed based on this environmental model to enable efficient and effective UAV path planning. Secondly, the bridge separation planning model is applied in a novel way, and a series of parameters are simulated, including the adjustment of the initial temperature value. The experimental results demonstrate that, compared with traditional local search algorithms, the proposed method achieves a cost reduction of 30.05\% and significantly improves effectiveness. Compared to path planning methods that do not consider wind factors, the proposed approach yields more realistic and practical results for UAV applications, as demonstrated by its improved effectiveness in simulations. These findings highlight the value of our method in facilitating more accurate and efficient UAV path planning in wind-prone environments.
comment: After carefully analysis, there is a bit design flaws in Algorithm 1. The experimental work of the paper is not comprehensive,which lacks an evaluation of the algorithm's running time
♻ ☆ Instance-aware Exploration-Verification-Exploitation for Instance ImageGoal Navigation
As a new embodied vision task, Instance ImageGoal Navigation (IIN) aims to navigate to a specified object depicted by a goal image in an unexplored environment. The main challenge of this task lies in identifying the target object from different viewpoints while rejecting similar distractors. Existing ImageGoal Navigation methods usually adopt the simple Exploration-Exploitation framework and ignore the identification of specific instance during navigation. In this work, we propose to imitate the human behaviour of ``getting closer to confirm" when distinguishing objects from a distance. Specifically, we design a new modular navigation framework named Instance-aware Exploration-Verification-Exploitation (IEVE) for instance-level image goal navigation. Our method allows for active switching among the exploration, verification, and exploitation actions, thereby facilitating the agent in making reasonable decisions under different situations. On the challenging HabitatMatterport 3D semantic (HM3D-SEM) dataset, our method surpasses previous state-of-the-art work, with a classical segmentation model (0.684 vs. 0.561 success) or a robust model (0.702 vs. 0.561 success)
♻ ☆ A Survey on Global LiDAR Localization: Challenges, Advances and Open Problems
Knowledge about the own pose is key for all mobile robot applications. Thus pose estimation is part of the core functionalities of mobile robots. Over the last two decades, LiDAR scanners have become the standard sensor for robot localization and mapping. This article aims to provide an overview of recent progress and advancements in LiDAR-based global localization. We begin by formulating the problem and exploring the application scope. We then present a review of the methodology, including recent advancements in several topics, such as maps, descriptor extraction, and cross-robot localization. The contents of the article are organized under three themes. The first theme concerns the combination of global place retrieval and local pose estimation. The second theme is upgrading single-shot measurements to sequential ones for sequential global localization. Finally, the third theme focuses on extending single-robot global localization to cross-robot localization in multi-robot systems. We conclude the survey with a discussion of open challenges and promising directions in global LiDAR localization. To our best knowledge, this is the first comprehensive survey on global LiDAR localization for mobile robots.
comment: Publishe on International Journal of Computer Vision (IJCV)
♻ ☆ Kinematic Modularity of Elementary Dynamic Actions
In this paper, a kinematically modular approach to robot control is presented. The method involves structures called Elementary Dynamic Actions and a network model combining these elements. With this control framework, a rich repertoire of movements can be generated by combination of basic modules. The problems of solving inverse kinematics, managing kinematic singularity and kinematic redundancy are avoided. The modular approach is robust against contact and physical interaction, which makes it particularly effective for contact-rich manipulation. Each kinematic module can be learned by Imitation Learning, thereby resulting in a modular learning strategy for robot control. The theoretical foundations and their real robot implementation are presented. Using a KUKA LBR iiwa14 robot, three tasks were considered: (1) generating a sequence of discrete movements, (2) generating a combination of discrete and rhythmic movements, and (3) a drawing and erasing task. The results obtained indicate that this modular approach has the potential to simplify the generation of a diverse range of robot actions.
comment: 8 pages, 4 figures
♻ ☆ Learning Hierarchical Control For Multi-Agent Capacity-Constrained Systems
This paper introduces a novel data-driven hierarchical control scheme for managing a fleet of nonlinear, capacity-constrained autonomous agents in an iterative environment. We propose a control framework consisting of a high-level dynamic task assignment and routing layer and low-level motion planning and tracking layer. Each layer of the control hierarchy uses a data-driven Model Predictive Control (MPC) policy, maintaining bounded computational complexity at each calculation of a new task assignment or actuation input. We utilize collected data to iteratively refine estimates of agent capacity usage, and update MPC policy parameters accordingly. Our approach leverages tools from iterative learning control to integrate learning at both levels of the hierarchy, and coordinates learning between levels in order to maintain closed-loop feasibility and performance improvement of the connected architecture.
♻ ☆ iSLAM: Imperative SLAM IEEE Robotics and Automation Letters
Simultaneous Localization and Mapping (SLAM) stands as one of the critical challenges in robot navigation. A SLAM system often consists of a front-end component for motion estimation and a back-end system for eliminating estimation drifts. Recent advancements suggest that data-driven methods are highly effective for front-end tasks, while geometry-based methods continue to be essential in the back-end processes. However, such a decoupled paradigm between the data-driven front-end and geometry-based back-end can lead to sub-optimal performance, consequently reducing the system's capabilities and generalization potential. To solve this problem, we proposed a novel self-supervised imperative learning framework, named imperative SLAM (iSLAM), which fosters reciprocal correction between the front-end and back-end, thus enhancing performance without necessitating any external supervision. Specifically, we formulate the SLAM problem as a bilevel optimization so that the front-end and back-end are bidirectionally connected. As a result, the front-end model can learn global geometric knowledge obtained through pose graph optimization by back-propagating the residuals from the back-end component. We showcase the effectiveness of this new framework through an application of stereo-inertial SLAM. The experiments show that the iSLAM training strategy achieves an accuracy improvement of 22% on average over a baseline model. To the best of our knowledge, iSLAM is the first SLAM system showing that the front-end and back-end components can mutually correct each other in a self-supervised manner.
comment: The paper has been accepted by IEEE Robotics and Automation Letters (RA-L)
♻ ☆ AutoTAMP: Autoregressive Task and Motion Planning with LLMs as Translators and Checkers
For effective human-robot interaction, robots need to understand, plan, and execute complex, long-horizon tasks described by natural language. Recent advances in large language models (LLMs) have shown promise for translating natural language into robot action sequences for complex tasks. However, existing approaches either translate the natural language directly into robot trajectories or factor the inference process by decomposing language into task sub-goals and relying on a motion planner to execute each sub-goal. When complex environmental and temporal constraints are involved, inference over planning tasks must be performed jointly with motion plans using traditional task-and-motion planning (TAMP) algorithms, making factorization into subgoals untenable. Rather than using LLMs to directly plan task sub-goals, we instead perform few-shot translation from natural language task descriptions to an intermediate task representation that can then be consumed by a TAMP algorithm to jointly solve the task and motion plan. To improve translation, we automatically detect and correct both syntactic and semantic errors via autoregressive re-prompting, resulting in significant improvements in task completion. We show that our approach outperforms several methods using LLMs as planners in complex task domains. See our project website https://yongchao98.github.io/MIT-REALM-AutoTAMP/ for prompts, videos, and code.
comment: 8 pages, 4 figures
♻ ☆ Scalable Multi-Robot Collaboration with Large Language Models: Centralized or Decentralized Systems?
A flurry of recent work has demonstrated that pre-trained large language models (LLMs) can be effective task planners for a variety of single-robot tasks. The planning performance of LLMs is significantly improved via prompting techniques, such as in-context learning or re-prompting with state feedback, placing new importance on the token budget for the context window. An under-explored but natural next direction is to investigate LLMs as multi-robot task planners. However, long-horizon, heterogeneous multi-robot planning introduces new challenges of coordination while also pushing up against the limits of context window length. It is therefore critical to find token-efficient LLM planning frameworks that are also able to reason about the complexities of multi-robot coordination. In this work, we compare the task success rate and token efficiency of four multi-agent communication frameworks (centralized, decentralized, and two hybrid) as applied to four coordination-dependent multi-agent 2D task scenarios for increasing numbers of agents. We find that a hybrid framework achieves better task success rates across all four tasks and scales better to more agents. We further demonstrate the hybrid frameworks in 3D simulations where the vision-to-text problem and dynamical errors are considered. See our project website https://yongchao98.github.io/MIT-REALM-Multi-Robot/ for prompts, videos, and code.
comment: 7 pages, 8 figures
♻ ☆ Learning Complex Motion Plans using Neural ODEs with Safety and Stability Guarantees ICRA 2024
We propose a Dynamical System (DS) approach to learn complex, possibly periodic motion plans from kinesthetic demonstrations using Neural Ordinary Differential Equations (NODE). To ensure reactivity and robustness to disturbances, we propose a novel approach that selects a target point at each time step for the robot to follow, by combining tools from control theory and the target trajectory generated by the learned NODE. A correction term to the NODE model is computed online by solving a quadratic program that guarantees stability and safety using control Lyapunov functions and control barrier functions, respectively. Our approach outperforms baseline DS learning techniques on the LASA handwriting dataset and complex periodic trajectories. It is also validated on the Franka Emika robot arm to produce stable motions for wiping and stirring tasks that do not have a single attractor, while being robust to perturbations and safe around humans and obstacles.
comment: accepted to ICRA 2024
♻ ☆ A Traffic Management Framework for On-Demand Urban Air Mobility Systems
Urban Air Mobility (UAM) offers a solution to current traffic congestion by providing on-demand air mobility in urban areas. Effective traffic management is crucial for efficient operation of UAM systems, especially for high-demand scenarios. In this paper, we present a centralized traffic management framework for on-demand UAM systems. Specifically, we provide a scheduling policy, called VertiSync, which schedules the aircraft for either servicing trip requests or rebalancing in the system subject to aircraft safety margins and energy requirements. We characterize the system-level throughput of VertiSync, which determines the demand threshold at which passenger waiting times transition from being stabilized to being increasing over time. We show that the proposed policy is able to maximize throughput for sufficiently large fleet sizes. We demonstrate the performance of VertiSync through a case study for the city of Los Angeles, and show that it significantly reduces passenger waiting times compared to a first-come first-serve scheduling policy.
comment: 9 pages, 6 figures
♻ ☆ Kinematics-aware Trajectory Generation and Prediction with Latent Stochastic Differential Modeling
Trajectory generation and trajectory prediction are two critical tasks in autonomous driving, which generate various trajectories for testing during development and predict the trajectories of surrounding vehicles during operation, respectively. In recent years, emerging data-driven deep learning-based methods have shown great promise for these two tasks in learning various traffic scenarios and improving average performance without assuming physical models. However, it remains a challenging problem for these methods to ensure that the generated/predicted trajectories are physically realistic. This challenge arises because learning-based approaches often function as opaque black boxes and do not adhere to physical laws. Conversely, existing model-based methods provide physically feasible results but are constrained by predefined model structures, limiting their capabilities to address complex scenarios. To address the limitations of these two types of approaches, we propose a new method that integrates kinematic knowledge into neural stochastic differential equations (SDE) and designs a variational autoencoder based on this latent kinematics-aware SDE (LK-SDE) to generate vehicle motions. Experimental results demonstrate that our method significantly outperforms both model-based and learning-based baselines in producing physically realistic and precisely controllable vehicle trajectories. Additionally, it performs well in predicting unobservable physical variables in the latent space.
comment: 8 pages, conference paper in motion generation
Artificial Intelligence 102
☆ LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models
Large Multimodal Models (LMMs) have shown significant reasoning capabilities by connecting a visual encoder and a large language model. LMMs typically use a fixed amount of visual tokens, such as the penultimate layer features in the CLIP visual encoder, as the prefix content. Recent LMMs incorporate more complex visual inputs, such as high-resolution images and videos, which increase the number of visual tokens significantly. However, due to the design of the Transformer architecture, computational costs associated with these models tend to increase quadratically with the number of input tokens. To tackle this problem, we explore a token reduction mechanism and find, similar to prior work, that many visual tokens are spatially redundant. Based on this, we propose PruMerge, a novel adaptive visual token reduction approach, which largely reduces the number of visual tokens while maintaining comparable model performance. We first select the unpruned visual tokens based on their similarity to class tokens and spatial tokens. We then cluster the pruned tokens based on key similarity and merge the clustered tokens with the unpruned tokens to supplement their information. Empirically, when applied to LLaVA-1.5, our approach can compress the visual tokens by 14.4 times on average, and achieve comparable performance across diverse visual question-answering and reasoning tasks. Code and checkpoints are at https://llava-prumerge.github.io/.
comment: Project page: https://llava-prumerge.github.io/
☆ LATTE3D: Large-scale Amortized Text-To-Enhanced3D Synthesis
Recent text-to-3D generation approaches produce impressive 3D results but require time-consuming optimization that can take up to an hour per prompt. Amortized methods like ATT3D optimize multiple prompts simultaneously to improve efficiency, enabling fast text-to-3D synthesis. However, they cannot capture high-frequency geometry and texture details and struggle to scale to large prompt sets, so they generalize poorly. We introduce LATTE3D, addressing these limitations to achieve fast, high-quality generation on a significantly larger prompt set. Key to our method is 1) building a scalable architecture and 2) leveraging 3D data during optimization through 3D-aware diffusion priors, shape regularization, and model initialization to achieve robustness to diverse and complex training prompts. LATTE3D amortizes both neural field and textured surface generation to produce highly detailed textured meshes in a single forward pass. LATTE3D generates 3D objects in 400ms, and can be further enhanced with fast test-time optimization.
comment: See the project website at https://research.nvidia.com/labs/toronto-ai/LATTE3D/
☆ Can large language models explore in-context?
We investigate the extent to which contemporary Large Language Models (LLMs) can engage in exploration, a core capability in reinforcement learning and decision making. We focus on native performance of existing LLMs, without training interventions. We deploy LLMs as agents in simple multi-armed bandit environments, specifying the environment description and interaction history entirely in-context, i.e., within the LLM prompt. We experiment with GPT-3.5, GPT-4, and Llama2, using a variety of prompt designs, and find that the models do not robustly engage in exploration without substantial interventions: i) Across all of our experiments, only one configuration resulted in satisfactory exploratory behavior: GPT-4 with chain-of-thought reasoning and an externally summarized interaction history, presented as sufficient statistics; ii) All other configurations did not result in robust exploratory behavior, including those with chain-of-thought reasoning but unsummarized history. Although these findings can be interpreted positively, they suggest that external summarization -- which may not be possible in more complex settings -- is important for obtaining desirable behavior from LLM agents. We conclude that non-trivial algorithmic interventions, such as fine-tuning or dataset curation, may be required to empower LLM-based decision making agents in complex settings.
☆ CoLLEGe: Concept Embedding Generation for Large Language Models
Current language models are unable to quickly learn new concepts on the fly, often requiring a more involved finetuning process to learn robustly. Prompting in-context is not robust to context distractions, and often fails to confer much information about the new concepts. Classic methods for few-shot word learning in NLP, relying on global word vectors, are less applicable to large language models. In this paper, we introduce a novel approach named CoLLEGe (Concept Learning with Language Embedding Generation) to modernize few-shot concept learning. CoLLEGe is a meta-learning framework capable of generating flexible embeddings for new concepts using a small number of example sentences or definitions. Our primary meta-learning objective is simply to facilitate a language model to make next word predictions in forthcoming sentences, making it compatible with language model pretraining. We design a series of tasks to test new concept learning in challenging real-world scenarios, including new word acquisition, definition inference, and verbal reasoning, and demonstrate that our method succeeds in each setting without task-specific training.
☆ Collaborative AI Teaming in Unknown Environments via Active Goal Deduction
With the advancements of artificial intelligence (AI), we're seeing more scenarios that require AI to work closely with other agents, whose goals and strategies might not be known beforehand. However, existing approaches for training collaborative agents often require defined and known reward signals and cannot address the problem of teaming with unknown agents that often have latent objectives/rewards. In response to this challenge, we propose teaming with unknown agents framework, which leverages kernel density Bayesian inverse learning method for active goal deduction and utilizes pre-trained, goal-conditioned policies to enable zero-shot policy adaptation. We prove that unbiased reward estimates in our framework are sufficient for optimal teaming with unknown agents. We further evaluate the framework of redesigned multi-agent particle and StarCraft II micromanagement environments with diverse unknown agents of different behaviors/rewards. Empirical results demonstrate that our framework significantly advances the teaming performance of AI and unknown agents in a wide range of collaborative scenarios.
☆ A Technological Perspective on Misuse of Available AI
Potential malicious misuse of civilian artificial intelligence (AI) poses serious threats to security on a national and international level. Besides defining autonomous systems from a technological viewpoint and explaining how AI development is characterized, we show how already existing and openly available AI technology could be misused. To underline this, we developed three exemplary use cases of potentially misused AI that threaten political, digital and physical security. The use cases can be built from existing AI technologies and components from academia, the private sector and the developer-community. This shows how freely available AI can be combined into autonomous weapon systems. Based on the use cases, we deduce points of control and further measures to prevent the potential threat through misused AI. Further, we promote the consideration of malicious misuse of civilian AI systems in the discussion on autonomous weapon systems (AWS).
comment: Presented at the UN Meeting of the Group of Governmental Experts on Lethal Autonomous Weapons Systems, 30 August 2018
☆ Point-DETR3D: Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection AAAI2024
Training high-accuracy 3D detectors necessitates massive labeled 3D annotations with 7 degree-of-freedom, which is laborious and time-consuming. Therefore, the form of point annotations is proposed to offer significant prospects for practical applications in 3D detection, which is not only more accessible and less expensive but also provides strong spatial information for object localization.In this paper, we empirically discover that it is non-trivial to merely adapt Point-DETR to its 3D form, encountering two main bottlenecks: 1) it fails to encode strong 3D prior into the model, and 2) it generates low-quality pseudo labels in distant regions due to the extreme sparsity of LiDAR points. To overcome these challenges, we introduce Point-DETR3D, a teacher-student framework for weakly semi-supervised 3D detection, designed to fully capitalize on point-wise supervision within a constrained instance-wise annotation budget.Different from Point-DETR which encodes 3D positional information solely through a point encoder, we propose an explicit positional query initialization strategy to enhance the positional prior. Considering the low quality of pseudo labels at distant regions produced by the teacher model, we enhance the detector's perception by incorporating dense imagery data through a novel Cross-Modal Deformable RoI Fusion (D-RoI).Moreover, an innovative point-guided self-supervised learning technique is proposed to allow for fully exploiting point priors, even in student models.Extensive experiments on representative nuScenes dataset demonstrate our Point-DETR3D obtains significant improvements compared to previous works. Notably, with only 5% of labeled data, Point-DETR3D achieves over 90% performance of its fully supervised counterpart.
comment: Accepted by AAAI2024
☆ CR3DT: Camera-RADAR Fusion for 3D Detection and Tracking
Accurate detection and tracking of surrounding objects is essential to enable self-driving vehicles. While Light Detection and Ranging (LiDAR) sensors have set the benchmark for high performance, the appeal of camera-only solutions lies in their cost-effectiveness. Notably, despite the prevalent use of Radio Detection and Ranging (RADAR) sensors in automotive systems, their potential in 3D detection and tracking has been largely disregarded due to data sparsity and measurement noise. As a recent development, the combination of RADARs and cameras is emerging as a promising solution. This paper presents Camera-RADAR 3D Detection and Tracking (CR3DT), a camera-RADAR fusion model for 3D object detection, and Multi-Object Tracking (MOT). Building upon the foundations of the State-of-the-Art (SotA) camera-only BEVDet architecture, CR3DT demonstrates substantial improvements in both detection and tracking capabilities, by incorporating the spatial and velocity information of the RADAR sensor. Experimental results demonstrate an absolute improvement in detection performance of 5.3% in mean Average Precision (mAP) and a 14.9% increase in Average Multi-Object Tracking Accuracy (AMOTA) on the nuScenes dataset when leveraging both modalities. CR3DT bridges the gap between high-performance and cost-effective perception systems in autonomous driving, by capitalizing on the ubiquitous presence of RADAR in automotive applications.
☆ KTbench: A Novel Data Leakage-Free Framework for Knowledge Tracing
Knowledge Tracing (KT) is concerned with predicting students' future performance on learning items in intelligent tutoring systems. Learning items are tagged with skill labels called knowledge concepts (KCs). Many KT models expand the sequence of item-student interactions into KC-student interactions by replacing learning items with their constituting KCs. This often results in a longer sequence length. This approach addresses the issue of sparse item-student interactions and minimises model parameters. However, two problems have been identified with such models. The first problem is the model's ability to learn correlations between KCs belonging to the same item, which can result in the leakage of ground truth labels and hinder performance. This problem can lead to a significant decrease in performance on datasets with a higher number of KCs per item. The second problem is that the available benchmark implementations ignore accounting for changes in sequence length when expanding KCs, leading to different models being tested with varying sequence lengths but still compared against the same benchmark. To address these problems, we introduce a general masking framework that mitigates the first problem and enhances the performance of such KT models while preserving the original model architecture without significant alterations. Additionally, we introduce KTbench, an open-source benchmark library designed to ensure the reproducibility of this work while mitigating the second problem.
comment: preprint
☆ Planning with a Learned Policy Basis to Optimally Solve Complex Tasks
Conventional reinforcement learning (RL) methods can successfully solve a wide range of sequential decision problems. However, learning policies that can generalize predictably across multiple tasks in a setting with non-Markovian reward specifications is a challenging problem. We propose to use successor features to learn a policy basis so that each (sub)policy in it solves a well-defined subproblem. In a task described by a finite state automaton (FSA) that involves the same set of subproblems, the combination of these (sub)policies can then be used to generate an optimal solution without additional learning. In contrast to other methods that combine (sub)policies via planning, our method asymptotically attains global optimality, even in stochastic environments.
☆ Sphere Neural-Networks for Rational Reasoning
The success of Large Language Models (LLMs), e.g., ChatGPT, is witnessed by their planetary popularity, their capability of human-like question-answering, and also by their steadily improved reasoning performance. However, it remains unclear whether LLMs reason. It is an open problem how traditional neural networks can be qualitatively extended to go beyond the statistic paradigm and achieve high-level cognition. Here, we present a minimalist qualitative extension by generalising computational building blocks from vectors to spheres. We propose Sphere Neural Networks (SphNNs) for human-like reasoning through model construction and inspection, and develop SphNN for syllogistic reasoning, a microcosm of human rationality. Instead of training data, SphNN uses a neuro-symbolic transition map of neighbourhood spatial relations to guide transformations from the current sphere configuration towards the target. SphNN is the first neural model that can determine the validity of long-chained syllogistic reasoning in one epoch by constructing sphere configurations as Euler diagrams, with the worst computational complexity of O(N^2). SphNN can evolve into various types of reasoning, such as spatio-temporal reasoning, logical reasoning with negation and disjunction, event reasoning, neuro-symbolic reasoning, and humour understanding (the highest level of cognition). All these suggest a new kind of Herbert A. Simon's scissors with two neural blades. SphNNs will tremendously enhance interdisciplinary collaborations to develop the two neural blades and realise deterministic neural reasoning and human-bounded rationality and elevate LLMs to reliable psychological AI. This work suggests that the non-zero radii of spheres are the missing components that prevent traditional deep-learning systems from reaching the realm of rational reasoning and cause LLMs to be trapped in the swamp of hallucination.
☆ Bioinformatics and Biomedical Informatics with ChatGPT: Year One Review
The year 2023 marked a significant surge in the exploration of applying large language model (LLM) chatbots, notably ChatGPT, across various disciplines. We surveyed the applications of ChatGPT in various sectors of bioinformatics and biomedical informatics throughout the year, covering omics, genetics, biomedical text mining, drug discovery, biomedical image understanding, bioinformatics programming, and bioinformatics education. Our survey delineates the current strengths and limitations of this chatbot in bioinformatics and offers insights into potential avenues for future development.
comment: 19 pages, 3 Figures, 1 Table
☆ Hierarchical Information Enhancement Network for Cascade Prediction in Social Networks
Understanding information cascades in networks is a fundamental issue in numerous applications. Current researches often sample cascade information into several independent paths or subgraphs to learn a simple cascade representation. However, these approaches fail to exploit the hierarchical semantic associations between different modalities, limiting their predictive performance. In this work, we propose a novel Hierarchical Information Enhancement Network (HIENet) for cascade prediction. Our approach integrates fundamental cascade sequence, user social graphs, and sub-cascade graph into a unified framework. Specifically, HIENet utilizes DeepWalk to sample cascades information into a series of sequences. It then gathers path information between users to extract the social relationships of propagators. Additionally, we employ a time-stamped graph convolutional network to aggregate sub-cascade graph information effectively. Ultimately, we introduce a Multi-modal Cascade Transformer to powerfully fuse these clues, providing a comprehensive understanding of cascading process. Extensive experiments have demonstrated the effectiveness of the proposed method.
comment: 7 pages, 2 figures
☆ Safe Learning of PDDL Domains with Conditional Effects -- Extended Version
Powerful domain-independent planners have been developed to solve various types of planning problems. These planners often require a model of the acting agent's actions, given in some planning domain description language. Manually designing such an action model is a notoriously challenging task. An alternative is to automatically learn action models from observation. Such an action model is called safe if every plan created with it is consistent with the real, unknown action model. Algorithms for learning such safe action models exist, yet they cannot handle domains with conditional or universal effects, which are common constructs in many planning problems. We prove that learning non-trivial safe action models with conditional effects may require an exponential number of samples. Then, we identify reasonable assumptions under which such learning is tractable and propose SAM Learning of Conditional Effects (Conditional-SAM), the first algorithm capable of doing so. We analyze Conditional-SAM theoretically and evaluate it experimentally. Our results show that the action models learned by Conditional-SAM can be used to solve perfectly most of the test set problems in most of the experimented domains.
☆ Comprehensive Reassessment of Large-Scale Evaluation Outcomes in LLMs: A Multifaceted Statistical Approach
Amidst the rapid evolution of LLMs, the significance of evaluation in comprehending and propelling these models forward is increasingly paramount. Evaluations have revealed that factors such as scaling, training types, architectures and other factors profoundly impact the performance of LLMs. However, the extent and nature of these impacts continue to be subjects of debate because most assessments have been restricted to a limited number of models and data points. Clarifying the effects of these factors on performance scores can be more effectively achieved through a statistical lens. Our study embarks on a thorough re-examination of these LLMs, targeting the inadequacies in current evaluation methods. With the advent of a uniform evaluation framework, our research leverages an expansive dataset of evaluation results, introducing a comprehensive statistical methodology. This includes the application of ANOVA, Tukey HSD tests, GAMM, and clustering technique, offering a robust and transparent approach to deciphering LLM performance data. Contrary to prevailing findings, our results challenge assumptions about emergent abilities and the influence of given training types and architectures in LLMs. These findings furnish new perspectives on the characteristics, intrinsic nature, and developmental trajectories of LLMs. By providing straightforward and reliable methods to scrutinize and reassess LLM performance data, this study contributes a nuanced perspective on LLM efficiency and potentials.
☆ Spectral Motion Alignment for Video Motion Transfer using Diffusion Models
The evolution of diffusion models has greatly impacted video generation and understanding. Particularly, text-to-video diffusion models (VDMs) have significantly facilitated the customization of input video with target appearance, motion, etc. Despite these advances, challenges persist in accurately distilling motion information from video frames. While existing works leverage the consecutive frame residual as the target motion vector, they inherently lack global motion context and are vulnerable to frame-wise distortions. To address this, we present Spectral Motion Alignment (SMA), a novel framework that refines and aligns motion vectors using Fourier and wavelet transforms. SMA learns motion patterns by incorporating frequency-domain regularization, facilitating the learning of whole-frame global motion dynamics, and mitigating spatial artifacts. Extensive experiments demonstrate SMA's efficacy in improving motion transfer while maintaining computational efficiency and compatibility across various video customization frameworks.
comment: Project page: https://geonyeong-park.github.io/spectral-motion-alignment/
Self-Supervised Backbone Framework for Diverse Agricultural Vision Tasks
Computer vision in agriculture is game-changing with its ability to transform farming into a data-driven, precise, and sustainable industry. Deep learning has empowered agriculture vision to analyze vast, complex visual data, but heavily rely on the availability of large annotated datasets. This remains a bottleneck as manual labeling is error-prone, time-consuming, and expensive. The lack of efficient labeling approaches inspired us to consider self-supervised learning as a paradigm shift, learning meaningful feature representations from raw agricultural image data. In this work, we explore how self-supervised representation learning unlocks the potential applicability to diverse agriculture vision tasks by eliminating the need for large-scale annotated datasets. We propose a lightweight framework utilizing SimCLR, a contrastive learning approach, to pre-train a ResNet-50 backbone on a large, unannotated dataset of real-world agriculture field images. Our experimental analysis and results indicate that the model learns robust features applicable to a broad range of downstream agriculture tasks discussed in the paper. Additionally, the reduced reliance on annotated data makes our approach more cost-effective and accessible, paving the way for broader adoption of computer vision in agriculture.
☆ Reasoning-Enhanced Object-Centric Learning for Videos
Object-centric learning aims to break down complex visual scenes into more manageable object representations, enhancing the understanding and reasoning abilities of machine learning systems toward the physical world. Recently, slot-based video models have demonstrated remarkable proficiency in segmenting and tracking objects, but they overlook the importance of the effective reasoning module. In the real world, reasoning and predictive abilities play a crucial role in human perception and object tracking; in particular, these abilities are closely related to human intuitive physics. Inspired by this, we designed a novel reasoning module called the Slot-based Time-Space Transformer with Memory buffer (STATM) to enhance the model's perception ability in complex scenes. The memory buffer primarily serves as storage for slot information from upstream modules, the Slot-based Time-Space Transformer makes predictions through slot-based spatiotemporal attention computations and fusion. Our experiment results on various datasets show that STATM can significantly enhance object-centric learning capabilities of slot-based video models.
☆ Multi-perspective Memory Enhanced Network for Identifying Key Nodes in Social Networks
Identifying key nodes in social networks plays a crucial role in timely blocking false information. Existing key node identification methods usually consider node influence only from the propagation structure perspective and have insufficient generalization ability to unknown scenarios. In this paper, we propose a novel Multi-perspective Memory Enhanced Network (MMEN) for identifying key nodes in social networks, which mines key nodes from multiple perspectives and utilizes memory networks to store historical information. Specifically, MMEN first constructs two propagation networks from the perspectives of user attributes and propagation structure and updates node feature representations using graph attention networks. Meanwhile, the memory network is employed to store information of similar subgraphs, enhancing the model's generalization performance in unknown scenarios. Finally, MMEN applies adaptive weights to combine the node influence of the two propagation networks to select the ultimate key nodes. Extensive experiments demonstrate that our method significantly outperforms previous methods.
comment: 7 pages, 1 figures
☆ Anytime, Anywhere, Anyone: Investigating the Feasibility of Segment Anything Model for Crowd-Sourcing Medical Image Annotations
Curating annotations for medical image segmentation is a labor-intensive and time-consuming task that requires domain expertise, resulting in "narrowly" focused deep learning (DL) models with limited translational utility. Recently, foundation models like the Segment Anything Model (SAM) have revolutionized semantic segmentation with exceptional zero-shot generalizability across various domains, including medical imaging, and hold a lot of promise for streamlining the annotation process. However, SAM has yet to be evaluated in a crowd-sourced setting to curate annotations for training 3D DL segmentation models. In this work, we explore the potential of SAM for crowd-sourcing "sparse" annotations from non-experts to generate "dense" segmentation masks for training 3D nnU-Net models, a state-of-the-art DL segmentation model. Our results indicate that while SAM-generated annotations exhibit high mean Dice scores compared to ground-truth annotations, nnU-Net models trained on SAM-generated annotations perform significantly worse than nnU-Net models trained on ground-truth annotations ($p<0.001$, all).
☆ (Un)making AI Magic: a Design Taxonomy
This paper examines the role that enchantment plays in the design of AI things by constructing a taxonomy of design approaches that increase or decrease the perception of magic and enchantment. We start from the design discourse surrounding recent developments in AI technologies, highlighting specific interaction qualities such as algorithmic uncertainties and errors and articulating relations to the rhetoric of magic and supernatural thinking. Through analyzing and reflecting upon 52 students' design projects from two editions of a Master course in design and AI, we identify seven design principles and unpack the effects of each in terms of enchantment and disenchantment. We conclude by articulating ways in which this taxonomy can be approached and appropriated by design/HCI practitioners, especially to support exploration and reflexivity.
☆ FSD-Inference: Fully Serverless Distributed Inference with Scalable Cloud Communication ICDE
Serverless computing offers attractive scalability, elasticity and cost-effectiveness. However, constraints on memory, CPU and function runtime have hindered its adoption for data-intensive applications and machine learning (ML) workloads. Traditional 'server-ful' platforms enable distributed computation via fast networks and well-established inter-process communication (IPC) mechanisms such as MPI and shared memory. In the absence of such solutions in the serverless domain, parallel computation with significant IPC requirements is challenging. We present FSD-Inference, the first fully serverless and highly scalable system for distributed ML inference. We explore potential communication channels, in conjunction with Function-as-a-Service (FaaS) compute, to design a state-of-the-art solution for distributed ML within the context of serverless data-intensive computing. We introduce novel fully serverless communication schemes for ML inference workloads, leveraging both cloud-based publish-subscribe/queueing and object storage offerings. We demonstrate how publish-subscribe/queueing services can be adapted for FaaS IPC with comparable performance to object storage, while offering significantly reduced cost at high parallelism levels. We conduct in-depth experiments on benchmark DNNs of various sizes. The results show that when compared to server-based alternatives, FSD-Inference is significantly more cost-effective and scalable, and can even achieve competitive performance against optimized HPC solutions. Experiments also confirm that our serverless solution can handle large distributed workloads and leverage high degrees of FaaS parallelism.
comment: In Proceedings of 2024 IEEE 40th International Conference on Data Engineering (ICDE) (to appear)
SFOD: Spiking Fusion Object Detector CVPR2024
Event cameras, characterized by high temporal resolution, high dynamic range, low power consumption, and high pixel bandwidth, offer unique capabilities for object detection in specialized contexts. Despite these advantages, the inherent sparsity and asynchrony of event data pose challenges to existing object detection algorithms. Spiking Neural Networks (SNNs), inspired by the way the human brain codes and processes information, offer a potential solution to these difficulties. However, their performance in object detection using event cameras is limited in current implementations. In this paper, we propose the Spiking Fusion Object Detector (SFOD), a simple and efficient approach to SNN-based object detection. Specifically, we design a Spiking Fusion Module, achieving the first-time fusion of feature maps from different scales in SNNs applied to event cameras. Additionally, through integrating our analysis and experiments conducted during the pretraining of the backbone network on the NCAR dataset, we delve deeply into the impact of spiking decoding strategies and loss functions on model performance. Thereby, we establish state-of-the-art classification results based on SNNs, achieving 93.7\% accuracy on the NCAR dataset. Experimental results on the GEN1 detection dataset demonstrate that the SFOD achieves a state-of-the-art mAP of 32.1\%, outperforming existing SNN-based approaches. Our research not only underscores the potential of SNNs in object detection with event cameras but also propels the advancement of SNNs. Code is available at https://github.com/yimeng-fan/SFOD.
comment: Accepted by CVPR2024
☆ Brain-grounding of semantic vectors improves neural decoding of visual stimuli
Developing algorithms for accurate and comprehensive neural decoding of mental contents is one of the long-cherished goals in the field of neuroscience and brain-machine interfaces. Previous studies have demonstrated the feasibility of neural decoding by training machine learning models to map brain activity patterns into a semantic vector representation of stimuli. These vectors, hereafter referred as pretrained feature vectors, are usually derived from semantic spaces based solely on image and/or text features and therefore they might have a totally different characteristics than how visual stimuli is represented in the human brain, resulting in limiting the capability of brain decoders to learn this mapping. To address this issue, we propose a representation learning framework, termed brain-grounding of semantic vectors, which fine-tunes pretrained feature vectors to better align with the neural representation of visual stimuli in the human brain. We trained this model this model with functional magnetic resonance imaging (fMRI) of 150 different visual stimuli categories, and then performed zero-shot brain decoding and identification analyses on 1) fMRI and 2) magnetoencephalography (MEG). Interestingly, we observed that by using the brain-grounded vectors, the brain decoding and identification accuracy on brain data from different neuroimaging modalities increases. These findings underscore the potential of incorporating a richer array of brain-derived features to enhance performance of brain decoding algorithms.
comment: 16 pages, 4 figures
☆ Exploring the Task-agnostic Trait of Self-supervised Learning in the Context of Detecting Mental Disorders
Self-supervised learning (SSL) has been investigated to generate task-agnostic representations across various domains. However, such investigation has not been conducted for detecting multiple mental disorders. The rationale behind the existence of a task-agnostic representation lies in the overlapping symptoms among multiple mental disorders. Consequently, the behavioural data collected for mental health assessment may carry a mixed bag of attributes related to multiple disorders. Motivated by that, in this study, we explore a task-agnostic representation derived through SSL in the context of detecting major depressive disorder (MDD) and post-traumatic stress disorder (PTSD) using audio and video data collected during interactive sessions. This study employs SSL models trained by predicting multiple fixed targets or masked frames. We propose a list of fixed targets to make the generated representation more efficient for detecting MDD and PTSD. Furthermore, we modify the hyper-parameters of the SSL encoder predicting fixed targets to generate global representations that capture varying temporal contexts. Both these innovations are noted to yield improved detection performances for considered mental disorders and exhibit task-agnostic traits. In the context of the SSL model predicting masked frames, the generated global representations are also noted to exhibit task-agnostic traits.
☆ Transition Graph Properties of Target Class Classification
Target class classification is a mixed classification and transition model whose integrated goal is to assign objects to a certain, so called target or normal class. The classification process is iterative, and in each step an object in a certain class undergoes an action attached to that class, initiating the transition of the object to one of the classes. The sequence of transitions, which we call class transitions, must be designed to provide the final assignment of objects to the target class. The transition process can be described in the form of a directed graph, and the success of the final classification is mainly due to the properties of this graph. In our previous research we showed that the desirable structure of the transition graph is an oriented rooted tree with orientation towards the root vertex, which corresponds to the normal class. It is clear that the transition graph of an arbitrary algorithm (policy) may not have this property. In this paper we study the structure of realistic transition graphs, which makes it possible to find classification inconsistencies, helping to transfer it into the desired form. The medical interpretation of dynamic treatment regime considered in the article further clarifies the investigated framework.
comment: 14pages, 4 figures
☆ Modular Deep Active Learning Framework for Image Annotation: A Technical Report for the Ophthalmo-AI Project
Image annotation is one of the most essential tasks for guaranteeing proper treatment for patients and tracking progress over the course of therapy in the field of medical imaging and disease diagnosis. However, manually annotating a lot of 2D and 3D imaging data can be extremely tedious. Deep Learning (DL) based segmentation algorithms have completely transformed this process and made it possible to automate image segmentation. By accurately segmenting medical images, these algorithms can greatly minimize the time and effort necessary for manual annotation. Additionally, by incorporating Active Learning (AL) methods, these segmentation algorithms can perform far more effectively with a smaller amount of ground truth data. We introduce MedDeepCyleAL, an end-to-end framework implementing the complete AL cycle. It provides researchers with the flexibility to choose the type of deep learning model they wish to employ and includes an annotation tool that supports the classification and segmentation of medical images. The user-friendly interface allows for easy alteration of the AL and DL model settings through a configuration file, requiring no prior programming experience. While MedDeepCyleAL can be applied to any kind of image data, we have specifically applied it to ophthalmology data in this project.
comment: DFKI Technical Report
☆ CACA Agent: Capability Collaboration based AI Agent
As AI Agents based on Large Language Models (LLMs) have shown potential in practical applications across various fields, how to quickly deploy an AI agent and how to conveniently expand the application scenario of AI agents has become a challenge. Previous studies mainly focused on implementing all the reasoning capabilities of AI agents within a single LLM, which often makes the model more complex and also reduces the extensibility of AI agent functionality. In this paper, we propose CACA Agent (Capability Collaboration based AI Agent), using an open architecture inspired by service computing. CACA Agent integrates a set of collaborative capabilities to implement AI Agents, not only reducing the dependence on a single LLM, but also enhancing the extensibility of both the planning abilities and the tools available to AI agents. Utilizing the proposed system, we present a demo to illustrate the operation and the application scenario extension of CACA Agent.
comment: 4 pages,5 figures
☆ Language Models in Dialogue: Conversational Maxims for Human-AI Interactions
Modern language models, while sophisticated, exhibit some inherent shortcomings, particularly in conversational settings. We claim that many of the observed shortcomings can be attributed to violation of one or more conversational principles. By drawing upon extensive research from both the social science and AI communities, we propose a set of maxims -- quantity, quality, relevance, manner, benevolence, and transparency -- for describing effective human-AI conversation. We first justify the applicability of the first four maxims (from Grice) in the context of human-AI interactions. We then argue that two new maxims, benevolence (concerning the generation of, and engagement with, harmful content) and transparency (concerning recognition of one's knowledge boundaries, operational constraints, and intents), are necessary for addressing behavior unique to modern human-AI interactions. The proposed maxims offer prescriptive guidance on how to assess conversational quality between humans and LLM-driven conversational agents, informing both their evaluation and improved design.
☆ Solving a Real-World Package Delivery Routing Problem Using Quantum Annealers
Research focused on the conjunction between quantum computing and routing problems has been very prolific in recent years. Most of the works revolve around classical problems such as the Traveling Salesman Problem or the Vehicle Routing Problem. Even though working on these problems is valuable, it is also undeniable that their academic-oriented nature falls short of real-world requirements. The main objective of this research is to present a solving method for realistic instances, avoiding problem relaxations or technical shortcuts. Instead, a quantum-classical hybrid solver has been developed, coined Q4RPD, that considers a set of real constraints such as a heterogeneous fleet of vehicles, priority deliveries, and capacities characterized by two values: weight and dimensions of the packages. Q4RPD resorts to the Leap Constrained Quadratic Model Hybrid Solver of D-Wave. To demonstrate the application of Q4RPD, an experimentation composed of six different instances has been conducted, aiming to serve as illustrative examples.
comment: 15 pages, 11 figures and 4 tables. Paper submitted for review in Scientific Reports
☆ Text clustering with LLM embeddings
Text clustering is an important approach for organising the growing amount of digital content, helping to structure and find hidden patterns in uncategorised data. In this research, we investigated how different textual embeddings - particularly those used in large language models (LLMs) - and clustering algorithms affect how text datasets are clustered. A series of experiments were conducted to assess how embeddings influence clustering results, the role played by dimensionality reduction through summarisation, and embedding size adjustment. Results reveal that LLM embeddings excel at capturing the nuances of structured language, while BERT leads the lightweight options in performance. In addition, we find that increasing embedding dimensionality and summarisation techniques do not uniformly improve clustering efficiency, suggesting that these strategies require careful analysis to use in real-life models. These results highlight a complex balance between the need for nuanced text representation and computational feasibility in text clustering applications. This study extends traditional text clustering frameworks by incorporating embeddings from LLMs, thereby paving the way for improved methodologies and opening new avenues for future research in various types of textual analysis.
☆ Subequivariant Reinforcement Learning Framework for Coordinated Motion Control
Effective coordination is crucial for motion control with reinforcement learning, especially as the complexity of agents and their motions increases. However, many existing methods struggle to account for the intricate dependencies between joints. We introduce CoordiGraph, a novel architecture that leverages subequivariant principles from physics to enhance coordination of motion control with reinforcement learning. This method embeds the principles of equivariance as inherent patterns in the learning process under gravity influence, which aids in modeling the nuanced relationships between joints vital for motion control. Through extensive experimentation with sophisticated agents in diverse environments, we highlight the merits of our approach. Compared to current leading methods, CoordiGraph notably enhances generalization and sample efficiency.
comment: 7 pages, 7 figures, 2024 IEEE International Conference on Robotics and Automation
☆ Argument-Aware Approach To Event Linking
Event linking connects event mentions in text with relevant nodes in a knowledge base (KB). Prior research in event linking has mainly borrowed methods from entity linking, overlooking the distinct features of events. Compared to the extensively explored entity linking task, events have more complex structures and can be more effectively distinguished by examining their associated arguments. Moreover, the information-rich nature of events leads to the scarcity of event KBs. This emphasizes the need for event linking models to identify and classify event mentions not in the KB as ``out-of-KB,'' an area that has received limited attention. In this work, we tackle these challenges by introducing an argument-aware approach. First, we improve event linking models by augmenting input text with tagged event argument information, facilitating the recognition of key information about event mentions. Subsequently, to help the model handle ``out-of-KB'' scenarios, we synthesize out-of-KB training examples from in-KB instances through controlled manipulation of event arguments. Our experiment across two test datasets showed significant enhancements in both in-KB and out-of-KB scenarios, with a notable 22% improvement in out-of-KB evaluations.
comment: Work In Progress
☆ Improved Long Short-Term Memory-based Wastewater Treatment Simulators for Deep Reinforcement Learning
Even though Deep Reinforcement Learning (DRL) showed outstanding results in the fields of Robotics and Games, it is still challenging to implement it in the optimization of industrial processes like wastewater treatment. One of the challenges is the lack of a simulation environment that will represent the actual plant as accurately as possible to train DRL policies. Stochasticity and non-linearity of wastewater treatment data lead to unstable and incorrect predictions of models over long time horizons. One possible reason for the models' incorrect simulation behavior can be related to the issue of compounding error, which is the accumulation of errors throughout the simulation. The compounding error occurs because the model utilizes its predictions as inputs at each time step. The error between the actual data and the prediction accumulates as the simulation continues. We implemented two methods to improve the trained models for wastewater treatment data, which resulted in more accurate simulators: 1- Using the model's prediction data as input in the training step as a tool of correction, and 2- Change in the loss function to consider the long-term predicted shape (dynamics). The experimental results showed that implementing these methods can improve the behavior of simulators in terms of Dynamic Time Warping throughout a year up to 98% compared to the base model. These improvements demonstrate significant promise in creating simulators for biological processes that do not need pre-existing knowledge of the process but instead depend exclusively on time series data obtained from the system.
☆ Comprehensive Lipidomic Automation Workflow using Large Language Models
Lipidomics generates large data that makes manual annotation and interpretation challenging. Lipid chemical and structural diversity with structural isomers further complicates annotation. Although, several commercial and open-source software for targeted lipid identification exists, it lacks automated method generation workflows and integration with statistical and bioinformatics tools. We have developed the Comprehensive Lipidomic Automated Workflow (CLAW) platform with integrated workflow for parsing, detailed statistical analysis and lipid annotations based on custom multiple reaction monitoring (MRM) precursor and product ion pair transitions. CLAW contains several modules including identification of carbon-carbon double bond position(s) in unsaturated lipids when combined with ozone electrospray ionization (OzESI)-MRM methodology. To demonstrate the utility of the automated workflow in CLAW, large-scale lipidomics data was collected with traditional and OzESI-MRM profiling on biological and non-biological samples. Specifically, a total of 1497 transitions organized into 10 MRM-based mass spectrometry methods were used to profile lipid droplets isolated from different brain regions of 18-24 month-old Alzheimer's disease mice and age-matched wild-type controls. Additionally, triacyclglycerols (TGs) profiles with carbon-carbon double bond specificity were generated from canola oil samples using OzESI-MRM profiling. We also developed an integrated language user interface with large language models using artificially intelligent (AI) agents that permits users to interact with the CLAW platform using a chatbot terminal to perform statistical and bioinformatic analyses. We envision CLAW pipeline to be used in high-throughput lipid structural identification tasks aiding users to generate automated lipidomics workflows ranging from data acquisition to AI agent-based bioinformatic analysis.
comment: 53 pages, 4 main figures, 23 Supporting figures, 10 Supporting Tables
☆ Bilateral Unsymmetrical Graph Contrastive Learning for Recommendation
Recent methods utilize graph contrastive Learning within graph-structured user-item interaction data for collaborative filtering and have demonstrated their efficacy in recommendation tasks. However, they ignore that the difference relation density of nodes between the user- and item-side causes the adaptability of graphs on bilateral nodes to be different after multi-hop graph interaction calculation, which limits existing models to achieve ideal results. To solve this issue, we propose a novel framework for recommendation tasks called Bilateral Unsymmetrical Graph Contrastive Learning (BusGCL) that consider the bilateral unsymmetry on user-item node relation density for sliced user and item graph reasoning better with bilateral slicing contrastive training. Especially, taking into account the aggregation ability of hypergraph-based graph convolutional network (GCN) in digging implicit similarities is more suitable for user nodes, embeddings generated from three different modules: hypergraph-based GCN, GCN and perturbed GCN, are sliced into two subviews by the user- and item-side respectively, and selectively combined into subview pairs bilaterally based on the characteristics of inter-node relation structure. Furthermore, to align the distribution of user and item embeddings after aggregation, a dispersing loss is leveraged to adjust the mutual distance between all embeddings for maintaining learning ability. Comprehensive experiments on two public datasets have proved the superiority of BusGCL in comparison to various recommendation methods. Other models can simply utilize our bilateral slicing contrastive learning to enhance recommending performance without incurring extra expenses.
☆ MM-Diff: High-Fidelity Image Personalization via Multi-Modal Condition Integration
Recent advances in tuning-free personalized image generation based on diffusion models are impressive. However, to improve subject fidelity, existing methods either retrain the diffusion model or infuse it with dense visual embeddings, both of which suffer from poor generalization and efficiency. Also, these methods falter in multi-subject image generation due to the unconstrained cross-attention mechanism. In this paper, we propose MM-Diff, a unified and tuning-free image personalization framework capable of generating high-fidelity images of both single and multiple subjects in seconds. Specifically, to simultaneously enhance text consistency and subject fidelity, MM-Diff employs a vision encoder to transform the input image into CLS and patch embeddings. CLS embeddings are used on the one hand to augment the text embeddings, and on the other hand together with patch embeddings to derive a small number of detail-rich subject embeddings, both of which are efficiently integrated into the diffusion model through the well-designed multimodal cross-attention mechanism. Additionally, MM-Diff introduces cross-attention map constraints during the training phase, ensuring flexible multi-subject image sampling during inference without any predefined inputs (e.g., layout). Extensive experiments demonstrate the superior performance of MM-Diff over other leading methods.
☆ Continual Vision-and-Language Navigation
Vision-and-Language Navigation (VLN) agents navigate to a destination using natural language instructions and the visual information they observe. Existing methods for training VLN agents presuppose fixed datasets, leading to a significant limitation: the introduction of new environments necessitates retraining with previously encountered environments to preserve their knowledge. This makes it difficult to train VLN agents that operate in the ever-changing real world. To address this limitation, we present the Continual Vision-and-Language Navigation (CVLN) paradigm, designed to evaluate agents trained through a continual learning process. For the training and evaluation of CVLN agents, we re-arrange existing VLN datasets to propose two datasets: CVLN-I, focused on navigation via initial-instruction interpretation, and CVLN-D, aimed at navigation through dialogue with other agents. Furthermore, we propose two novel rehearsal-based methods for CVLN, Perplexity Replay (PerpR) and Episodic Self-Replay (ESR). PerpR prioritizes replaying challenging episodes based on action perplexity, while ESR replays previously predicted action logits to preserve learned behaviors. We demonstrate the effectiveness of the proposed methods on CVLN through extensive experiments.
☆ Cartoon Hallucinations Detection: Pose-aware In Context Visual Learning
Large-scale Text-to-Image (TTI) models have become a common approach for generating training data in various generative fields. However, visual hallucinations, which contain perceptually critical defects, remain a concern, especially in non-photorealistic styles like cartoon characters. We propose a novel visual hallucination detection system for cartoon character images generated by TTI models. Our approach leverages pose-aware in-context visual learning (PA-ICVL) with Vision-Language Models (VLMs), utilizing both RGB images and pose information. By incorporating pose guidance from a fine-tuned pose estimator, we enable VLMs to make more accurate decisions. Experimental results demonstrate significant improvements in identifying visual hallucinations compared to baseline methods relying solely on RGB images. This research advances TTI models by mitigating visual hallucinations, expanding their potential in non-photorealistic domains.
comment: 11 pages, 12 figures, 1 table, Project page: https://gh-bumsookim.github.io/Cartoon-Hallucinations-Detection/
☆ Multimodal Fusion with Pre-Trained Model Features in Affective Behaviour Analysis In-the-wild
Multimodal fusion is a significant method for most multimodal tasks. With the recent surge in the number of large pre-trained models, combining both multimodal fusion methods and pre-trained model features can achieve outstanding performance in many multimodal tasks. In this paper, we present our approach, which leverages both advantages for addressing the task of Expression (Expr) Recognition and Valence-Arousal (VA) Estimation. We evaluate the Aff-Wild2 database using pre-trained models, then extract the final hidden layers of the models as features. Following preprocessing and interpolation or convolution to align the extracted features, different models are employed for modal fusion. Our code is available at GitHub - FulgenceWen/ABAW6th.
☆ Grey-informed neural network for time-series forecasting
Neural network models have shown outstanding performance and successful resolutions to complex problems in various fields. However, the majority of these models are viewed as black-box, requiring a significant amount of data for development. Consequently, in situations with limited data, constructing appropriate models becomes challenging due to the lack of transparency and scarcity of data. To tackle these challenges, this study suggests the implementation of a grey-informed neural network (GINN). The GINN ensures that the output of the neural network follows the differential equation model of the grey system, improving interpretability. Moreover, incorporating prior knowledge from grey system theory enables traditional neural networks to effectively handle small data samples. Our proposed model has been observed to uncover underlying patterns in the real world and produce reliable forecasts based on empirical data.
☆ Magic for the Age of Quantized DNNs
Recently, the number of parameters in DNNs has explosively increased, as exemplified by LLMs (Large Language Models), making inference on small-scale computers more difficult. Model compression technology is, therefore, essential for integration into products. In this paper, we propose a method of quantization-aware training. We introduce a novel normalization (Layer-Batch Normalization) that is independent of the mini-batch size and does not require any additional computation cost during inference. Then, we quantize the weights by the scaled round-clip function with the weight standardization. We also quantize activation functions using the same function and apply surrogate gradients to train the model with both quantized weights and the quantized activation functions. We call this method Magic for the age of Quantised DNNs (MaQD). Experimental results show that our quantization method can be achieved with minimal accuracy degradation.
comment: 14 pages, 5 figures, 4 tables
☆ Piecewise-Linear Manifolds for Deep Metric Learning
Unsupervised deep metric learning (UDML) focuses on learning a semantic representation space using only unlabeled data. This challenging problem requires accurately estimating the similarity between data points, which is used to supervise a deep network. For this purpose, we propose to model the high-dimensional data manifold using a piecewise-linear approximation, with each low-dimensional linear piece approximating the data manifold in a small neighborhood of a point. These neighborhoods are used to estimate similarity between data points. We empirically show that this similarity estimate correlates better with the ground truth than the similarity estimates of current state-of-the-art techniques. We also show that proxies, commonly used in supervised metric learning, can be used to model the piecewise-linear manifold in an unsupervised setting, helping improve performance. Our method outperforms existing unsupervised metric learning approaches on standard zero-shot image retrieval benchmarks.
comment: Accepted at CPAL 2024 (Oral)
☆ A Picture Is Worth a Graph: Blueprint Debate on Graph for Multimodal Reasoning
This paper presents a pilot study aimed at introducing multi-agent debate into multimodal reasoning. The study addresses two key challenges: the trivialization of opinions resulting from excessive summarization and the diversion of focus caused by distractor concepts introduced from images. These challenges stem from the inductive (bottom-up) nature of existing debating schemes. To address the issue, we propose a deductive (top-down) debating approach called Blueprint Debate on Graphs (BDoG). In BDoG, debates are confined to a blueprint graph to prevent opinion trivialization through world-level summarization. Moreover, by storing evidence in branches within the graph, BDoG mitigates distractions caused by frequent but irrelevant concepts. Extensive experiments validate BDoG, achieving state-of-the-art results in Science QA and MMBench with significant improvements over previous methods.
comment: Work in progress
☆ Comprehensive Evaluation and Insights into the Use of Large Language Models in the Automation of Behavior-Driven Development Acceptance Test Formulation
Behavior-driven development (BDD) is an Agile testing methodology fostering collaboration among developers, QA analysts, and stakeholders. In this manuscript, we propose a novel approach to enhance BDD practices using large language models (LLMs) to automate acceptance test generation. Our study uses zero and few-shot prompts to evaluate LLMs such as GPT-3.5, GPT-4, Llama-2-13B, and PaLM-2. The paper presents a detailed methodology that includes the dataset, prompt techniques, LLMs, and the evaluation process. The results demonstrate that GPT-3.5 and GPT-4 generate error-free BDD acceptance tests with better performance. The few-shot prompt technique highlights its ability to provide higher accuracy by incorporating examples for in-context learning. Furthermore, the study examines syntax errors, validation accuracy, and comparative analysis of LLMs, revealing their effectiveness in enhancing BDD practices. However, our study acknowledges that there are limitations to the proposed approach. We emphasize that this approach can support collaborative BDD processes and create opportunities for future research into automated BDD acceptance test generation using LLMs.
☆ Evidence-Driven Retrieval Augmented Response Generation for Online Misinformation NAACL 2024
The proliferation of online misinformation has posed significant threats to public interest. While numerous online users actively participate in the combat against misinformation, many of such responses can be characterized by the lack of politeness and supporting facts. As a solution, text generation approaches are proposed to automatically produce counter-misinformation responses. Nevertheless, existing methods are often trained end-to-end without leveraging external knowledge, resulting in subpar text quality and excessively repetitive responses. In this paper, we propose retrieval augmented response generation for online misinformation (RARG), which collects supporting evidence from scientific sources and generates counter-misinformation responses based on the evidences. In particular, our RARG consists of two stages: (1) evidence collection, where we design a retrieval pipeline to retrieve and rerank evidence documents using a database comprising over 1M academic articles; (2) response generation, in which we align large language models (LLMs) to generate evidence-based responses via reinforcement learning from human feedback (RLHF). We propose a reward function to maximize the utilization of the retrieved evidence while maintaining the quality of the generated text, which yields polite and factual responses that clearly refutes misinformation. To demonstrate the effectiveness of our method, we study the case of COVID-19 and perform extensive experiments with both in- and cross-domain datasets, where RARG consistently outperforms baselines by generating high-quality counter-misinformation responses.
comment: Accepted to NAACL 2024
☆ Simple Graph Condensation
The burdensome training costs on large-scale graphs have aroused significant interest in graph condensation, which involves tuning Graph Neural Networks (GNNs) on a small condensed graph for use on the large-scale original graph. Existing methods primarily focus on aligning key metrics between the condensed and original graphs, such as gradients, distribution and trajectory of GNNs, yielding satisfactory performance on downstream tasks. However, these complex metrics necessitate intricate computations and can potentially disrupt the optimization process of the condensation graph, making the condensation process highly demanding and unstable. Motivated by the recent success of simplified models in various fields, we propose a simplified approach to metric alignment in graph condensation, aiming to reduce unnecessary complexity inherited from GNNs. In our approach, we eliminate external parameters and exclusively retain the target condensed graph during the condensation process. Following the hierarchical aggregation principles of GNNs, we introduce the Simple Graph Condensation (SimGC) framework, which aligns the condensed graph with the original graph from the input layer to the prediction layer, guided by a pre-trained Simple Graph Convolution (SGC) model on the original graph. As a result, both graphs possess the similar capability to train GNNs. This straightforward yet effective strategy achieves a significant speedup of up to 10 times compared to existing graph condensation methods while performing on par with state-of-the-art baselines. Comprehensive experiments conducted on seven benchmark datasets demonstrate the effectiveness of SimGC in prediction accuracy, condensation time, and generalization capability. Our code will be made publicly available.
comment: Under review
☆ A Single Linear Layer Yields Task-Adapted Low-Rank Matrices LREC
Low-Rank Adaptation (LoRA) is a widely used Parameter-Efficient Fine-Tuning (PEFT) method that updates an initial weight matrix $W_0$ with a delta matrix $\Delta W$ consisted by two low-rank matrices $A$ and $B$. A previous study suggested that there is correlation between $W_0$ and $\Delta W$. In this study, we aim to delve deeper into relationships between $W_0$ and low-rank matrices $A$ and $B$ to further comprehend the behavior of LoRA. In particular, we analyze a conversion matrix that transform $W_0$ into low-rank matrices, which encapsulates information about the relationships. Our analysis reveals that the conversion matrices are similar across each layer. Inspired by these findings, we hypothesize that a single linear layer, which takes each layer's $W_0$ as input, can yield task-adapted low-rank matrices. To confirm this hypothesis, we devise a method named Conditionally Parameterized LoRA (CondLoRA) that updates initial weight matrices with low-rank matrices derived from a single linear layer. Our empirical results show that CondLoRA maintains a performance on par with LoRA, despite the fact that the trainable parameters of CondLoRA are fewer than those of LoRA. Therefore, we conclude that "a single linear layer yields task-adapted low-rank matrices."
comment: Accepted at LREC-COLING 2024
☆ Unifying Lane-Level Traffic Prediction from a Graph Structural Perspective: Benchmark and Baseline
Traffic prediction has long been a focal and pivotal area in research, witnessing both significant strides from city-level to road-level predictions in recent years. With the advancement of Vehicle-to-Everything (V2X) technologies, autonomous driving, and large-scale models in the traffic domain, lane-level traffic prediction has emerged as an indispensable direction. However, further progress in this field is hindered by the absence of comprehensive and unified evaluation standards, coupled with limited public availability of data and code. This paper extensively analyzes and categorizes existing research in lane-level traffic prediction, establishes a unified spatial topology structure and prediction tasks, and introduces a simple baseline model, GraphMLP, based on graph structure and MLP networks. We have replicated codes not publicly available in existing studies and, based on this, thoroughly and fairly assessed various models in terms of effectiveness, efficiency, and applicability, providing insights for practical applications. Additionally, we have released three new datasets and corresponding codes to accelerate progress in this field, all of which can be found on https://github.com/ShuhaoLii/TITS24LaneLevel-Traffic-Benchmark.
☆ Attention-Driven Reasoning: Unlocking the Potential of Large Language Models
Large Language Models (LLMs) have shown remarkable capabilities, but their reasoning abilities and underlying mechanisms remain poorly understood. We present a novel approach to enhance LLMs' reasoning through attention mechanism optimization, without additional training data. We identify inefficiencies in the attention distribution caused by non-semantic tokens and propose an algorithm to re-balance the skewed distribution, enabling the model to abstract more nuanced knowledge. Our experiments demonstrate significantly improved reasoning capabilities, particularly for non-STEM questions. We provide insights into the role of attention patterns in LLMs' reasoning and propose a method to enhance these abilities, paving the way for more powerful and versatile language models.
☆ Hierarchical Skip Decoding for Efficient Autoregressive Text Generation
Autoregressive decoding strategy is a commonly used method for text generation tasks with pre-trained language models, while early-exiting is an effective approach to speedup the inference stage. In this work, we propose a novel decoding strategy named Hierarchical Skip Decoding (HSD) for efficient autoregressive text generation. Different from existing methods that require additional trainable components, HSD is a plug-and-play method applicable to autoregressive text generation models, it adaptively skips decoding layers in a hierarchical manner based on the current sequence length, thereby reducing computational workload and allocating computation resources. Comprehensive experiments on five text generation datasets with pre-trained language models demonstrate HSD's advantages in balancing efficiency and text quality. With almost half of the layers skipped, HSD can sustain 90% of the text quality compared to vanilla autoregressive decoding, outperforming the competitive approaches.
☆ Stance Reasoner: Zero-Shot Stance Detection on Social Media with Explicit Reasoning COLING 2024
Social media platforms are rich sources of opinionated content. Stance detection allows the automatic extraction of users' opinions on various topics from such content. We focus on zero-shot stance detection, where the model's success relies on (a) having knowledge about the target topic; and (b) learning general reasoning strategies that can be employed for new topics. We present Stance Reasoner, an approach to zero-shot stance detection on social media that leverages explicit reasoning over background knowledge to guide the model's inference about the document's stance on a target. Specifically, our method uses a pre-trained language model as a source of world knowledge, with the chain-of-thought in-context learning approach to generate intermediate reasoning steps. Stance Reasoner outperforms the current state-of-the-art models on 3 Twitter datasets, including fully supervised models. It can better generalize across targets, while at the same time providing explicit and interpretable explanations for its predictions.
comment: Accepted to COLING 2024
♻ ☆ Simple and Scalable Strategies to Continually Pre-train Large Language Models
Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant compute compared to re-training. However, the distribution shift induced by new data typically results in degraded performance on previous data or poor adaptation to the new data. In this work, we show that a simple and scalable combination of learning rate (LR) re-warming, LR re-decaying, and replay of previous data is sufficient to match the performance of fully re-training from scratch on all available data, as measured by final loss and language model (LM) evaluation benchmarks. Specifically, we show this for a weak but realistic distribution shift between two commonly used LLM pre-training datasets (English$\rightarrow$English) and a stronger distribution shift (English$\rightarrow$German) at the $405$M parameter model scale with large dataset sizes (hundreds of billions of tokens). Selecting the weak but realistic shift for larger-scale experiments, we also find that our continual learning strategies match the re-training baseline for a 10B parameter LLM. Our results demonstrate that LLMs can be successfully updated via simple and scalable continual learning strategies, matching the re-training baseline using only a fraction of the compute. Finally, inspired by previous work, we propose alternatives to the cosine learning rate schedule that help circumvent forgetting induced by LR re-warming and that are not bound to a fixed token budget.
♻ ☆ Finding the right XAI method -- A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Science
Explainable artificial intelligence (XAI) methods shed light on the predictions of machine learning algorithms. Several different approaches exist and have already been applied in climate science. However, usually missing ground truth explanations complicate their evaluation and comparison, subsequently impeding the choice of the XAI method. Therefore, in this work, we introduce XAI evaluation in the climate context and discuss different desired explanation properties, namely robustness, faithfulness, randomization, complexity, and localization. To this end, we chose previous work as a case study where the decade of annual-mean temperature maps is predicted. After training both a multi-layer perceptron (MLP) and a convolutional neural network (CNN), multiple XAI methods are applied and their skill scores in reference to a random uniform explanation are calculated for each property. Independent of the network, we find that XAI methods Integrated Gradients, layer-wise relevance propagation, and input times gradients exhibit considerable robustness, faithfulness, and complexity while sacrificing randomization performance. Sensitivity methods -- gradient, SmoothGrad, NoiseGrad, and FusionGrad, match the robustness skill but sacrifice faithfulness and complexity for randomization skill. We find architecture-dependent performance differences regarding robustness, complexity and localization skills of different XAI methods, highlighting the necessity for research task-specific evaluation. Overall, our work offers an overview of different evaluation properties in the climate science context and shows how to compare and benchmark different explanation methods, assessing their suitability based on strengths and weaknesses, for the specific research problem at hand. By that, we aim to support climate researchers in the selection of a suitable XAI method.
comment: 19 pages, 10 figure, accepted at AIES journal by AMS
♻ ☆ Videoshop: Localized Semantic Video Editing with Noise-Extrapolated Diffusion Inversion
We introduce Videoshop, a training-free video editing algorithm for localized semantic edits. Videoshop allows users to use any editing software, including Photoshop and generative inpainting, to modify the first frame; it automatically propagates those changes, with semantic, spatial, and temporally consistent motion, to the remaining frames. Unlike existing methods that enable edits only through imprecise textual instructions, Videoshop allows users to add or remove objects, semantically change objects, insert stock photos into videos, etc. with fine-grained control over locations and appearance. We achieve this through image-based video editing by inverting latents with noise extrapolation, from which we generate videos conditioned on the edited image. Videoshop produces higher quality edits against 6 baselines on 2 editing benchmarks using 10 evaluation metrics.
comment: Project page at https://videoshop-editing.github.io/
♻ ☆ Empowering Autonomous Driving with Large Language Models: A Safety Perspective ICLR2024
Autonomous Driving (AD) encounters significant safety hurdles in long-tail unforeseen driving scenarios, largely stemming from the non-interpretability and poor generalization of the deep neural networks within the AD system, particularly in out-of-distribution and uncertain data. To this end, this paper explores the integration of Large Language Models (LLMs) into AD systems, leveraging their robust common-sense knowledge and reasoning abilities. The proposed methodologies employ LLMs as intelligent decision-makers in behavioral planning, augmented with a safety verifier shield for contextual safety learning, for enhancing driving performance and safety. We present two key studies in a simulated environment: an adaptive LLM-conditioned Model Predictive Control (MPC) and an LLM-enabled interactive behavior planning scheme with a state machine. Demonstrating superior performance and safety metrics compared to state-of-the-art approaches, our approach shows the promising potential for using LLMs for autonomous vehicles.
comment: Accepted to LLMAgent workshop @ICLR2024
♻ ☆ LLMR: Real-time Prompting of Interactive Worlds using Large Language Models
We present Large Language Model for Mixed Reality (LLMR), a framework for the real-time creation and modification of interactive Mixed Reality experiences using LLMs. LLMR leverages novel strategies to tackle difficult cases where ideal training data is scarce, or where the design goal requires the synthesis of internal dynamics, intuitive analysis, or advanced interactivity. Our framework relies on text interaction and the Unity game engine. By incorporating techniques for scene understanding, task planning, self-debugging, and memory management, LLMR outperforms the standard GPT-4 by 4x in average error rate. We demonstrate LLMR's cross-platform interoperability with several example worlds, and evaluate it on a variety of creation and modification tasks to show that it can produce and edit diverse objects, tools, and scenes. Finally, we conducted a usability study (N=11) with a diverse set that revealed participants had positive experiences with the system and would use it again.
comment: 46 pages, 18 figures; Matching version accepted at CHI 2024
♻ ☆ From Complexity to Clarity: Analytical Expressions of Deep Neural Network Weights via Clifford's Geometric Algebra and Convexity
In this paper, we introduce a novel analysis of neural networks based on geometric (Clifford) algebra and convex optimization. We show that optimal weights of deep ReLU neural networks are given by the wedge product of training samples when trained with standard regularized loss. Furthermore, the training problem reduces to convex optimization over wedge product features, which encode the geometric structure of the training dataset. This structure is given in terms of signed volumes of triangles and parallelotopes generated by data vectors. The convex problem finds a small subset of samples via $\ell_1$ regularization to discover only relevant wedge product features. Our analysis provides a novel perspective on the inner workings of deep neural networks and sheds light on the role of the hidden layers.
♻ ☆ Building Efficient Universal Classifiers with Natural Language Inference
Generative Large Language Models (LLMs) have become the mainstream choice for fewshot and zeroshot learning thanks to the universality of text generation. Many users, however, do not need the broad capabilities of generative LLMs when they only want to automate a classification task. Smaller BERT-like models can also learn universal tasks, which allow them to do any text classification task without requiring fine-tuning (zeroshot classification) or to learn new tasks with only a few examples (fewshot), while being significantly more efficient than generative LLMs. This paper (1) explains how Natural Language Inference (NLI) can be used as a universal classification task that follows similar principles as instruction fine-tuning of generative LLMs, (2) provides a step-by-step guide with reusable Jupyter notebooks for building a universal classifier, and (3) shares the resulting universal classifier that is trained on 33 datasets with 389 diverse classes. Parts of the code we share has been used to train our older zeroshot classifiers that have been downloaded more than 55 million times via the Hugging Face Hub as of December 2023. Our new classifier improves zeroshot performance by 9.4%.
♻ ☆ VideoPoet: A Large Language Model for Zero-Shot Video Generation
We present VideoPoet, a language model capable of synthesizing high-quality video, with matching audio, from a large variety of conditioning signals. VideoPoet employs a decoder-only transformer architecture that processes multimodal inputs -- including images, videos, text, and audio. The training protocol follows that of Large Language Models (LLMs), consisting of two stages: pretraining and task-specific adaptation. During pretraining, VideoPoet incorporates a mixture of multimodal generative objectives within an autoregressive Transformer framework. The pretrained LLM serves as a foundation that can be adapted for a range of video generation tasks. We present empirical results demonstrating the model's state-of-the-art capabilities in zero-shot video generation, specifically highlighting VideoPoet's ability to generate high-fidelity motions. Project page: http://sites.research.google/videopoet/
comment: Project page: http://sites.research.google/videopoet/
♻ ☆ Fast ODE-based Sampling for Diffusion Models in Around 5 Steps CVPR 2024
Sampling from diffusion models can be treated as solving the corresponding ordinary differential equations (ODEs), with the aim of obtaining an accurate solution with as few number of function evaluations (NFE) as possible. Recently, various fast samplers utilizing higher-order ODE solvers have emerged and achieved better performance than the initial first-order one. However, these numerical methods inherently result in certain approximation errors, which significantly degrades sample quality with extremely small NFE (e.g., around 5). In contrast, based on the geometric observation that each sampling trajectory almost lies in a two-dimensional subspace embedded in the ambient space, we propose Approximate MEan-Direction Solver (AMED-Solver) that eliminates truncation errors by directly learning the mean direction for fast diffusion sampling. Besides, our method can be easily used as a plugin to further improve existing ODE-based samplers. Extensive experiments on image synthesis with the resolution ranging from 32 to 512 demonstrate the effectiveness of our method. With only 5 NFE, we achieve 6.61 FID on CIFAR-10, 10.74 FID on ImageNet 64$\times$64, and 13.20 FID on LSUN Bedroom. Our code is available at https://github.com/zju-pi/diff-sampler.
comment: Accepted by CVPR 2024
♻ ☆ Novelty Detection in Reinforcement Learning with World Models
Reinforcement learning (RL) using world models has found significant recent successes. However, when a sudden change to world mechanics or properties occurs then agent performance and reliability can dramatically decline. We refer to the sudden change in visual properties or state transitions as novelties. Implementing novelty detection within generated world model frameworks is a crucial task for protecting the agent when deployed. In this paper, we propose straightforward bounding approaches to incorporate novelty detection into world model RL agents, by utilizing the misalignment of the world model's hallucinated states and the true observed states as an anomaly score. We provide effective approaches to detecting novelties in a distribution of transitions learned by an agent in a world model. Finally, we show the advantage of our work in a novel environment compared to traditional machine learning novelty detection methods as well as currently accepted RL focused novelty detection algorithms.
♻ ☆ Faster Neighborhood Attention: Reducing the O(n^2) Cost of Self Attention at the Threadblock Level
Neighborhood attention reduces the cost of self attention by restricting each token's attention span to its nearest neighbors. This restriction, parameterized by a window size and dilation factor, draws a spectrum of possible attention patterns between linear projection and self attention. Neighborhood attention, and more generally sliding window attention patterns, have long been bounded by infrastructure, particularly in higher-rank spaces (2-D and 3-D), calling for the development of custom kernels, which have been limited in either functionality, or performance, if not both. In this work, we first show that neighborhood attention can be represented as a batched GEMM problem, similar to standard attention, and implement it for 1-D and 2-D neighborhood attention. These kernels on average provide 895% and 272% improvement in full precision latency compared to existing naive kernels for 1-D and 2-D neighborhood attention respectively. We find certain inherent inefficiencies in all unfused neighborhood attention kernels that bound their performance and lower-precision scalability. We also developed fused neighborhood attention; an adaptation of fused dot-product attention kernels that allow fine-grained control over attention across different spatial axes. Known for reducing the quadratic time complexity of self attention to a linear complexity, neighborhood attention can now enjoy a reduced and constant memory footprint, and record-breaking half precision latency. We observe that our fused kernels successfully circumvent some of the unavoidable inefficiencies in unfused implementations. While our unfused GEMM-based kernels only improve half precision performance compared to naive kernels by an average of 496% and 113% in 1-D and 2-D problems respectively, our fused kernels improve naive kernels by an average of 1607% and 581% in 1-D and 2-D problems respectively.
comment: Project page: https://github.com/SHI-Labs/NATTEN
♻ ☆ Large Language Model-informed ECG Dual Attention Network for Heart Failure Risk Prediction
Heart failure (HF) poses a significant public health challenge, with a rising global mortality rate. Early detection and prevention of HF could significantly reduce its impact. We introduce a novel methodology for predicting HF risk using 12-lead electrocardiograms (ECGs). We present a novel, lightweight dual-attention ECG network designed to capture complex ECG features essential for early HF risk prediction, despite the notable imbalance between low and high-risk groups. This network incorporates a cross-lead attention module and twelve lead-specific temporal attention modules, focusing on cross-lead interactions and each lead's local dynamics. To further alleviate model overfitting, we leverage a large language model (LLM) with a public ECG-Report dataset for pretraining on an ECG-report alignment task. The network is then fine-tuned for HF risk prediction using two specific cohorts from the UK Biobank study, focusing on patients with hypertension (UKB-HYP) and those who have had a myocardial infarction (UKB-MI).The results reveal that LLM-informed pre-training substantially enhances HF risk prediction in these cohorts. The dual-attention design not only improves interpretability but also predictive accuracy, outperforming existing competitive methods with C-index scores of 0.6349 for UKB-HYP and 0.5805 for UKB-MI. This demonstrates our method's potential in advancing HF risk assessment with clinical complex ECG data.
comment: Under journal revision
♻ ☆ "This is not a data problem": Algorithms and Power in Public Higher Education in Canada
Algorithmic decision-making is increasingly being adopted across public higher education. The expansion of data-driven practices by post-secondary institutions has occurred in parallel with the adoption of New Public Management approaches by neoliberal administrations. In this study, we conduct a qualitative analysis of an in-depth ethnographic case study of data and algorithms in use at a public college in Ontario, Canada. We identify the data, algorithms, and outcomes in use at the college. We assess how the college's processes and relationships support those outcomes and the different stakeholders' perceptions of the college's data-driven systems. In addition, we find that the growing reliance on algorithmic decisions leads to increased student surveillance, exacerbation of existing inequities, and the automation of the faculty-student relationship. Finally, we identify a cycle of increased institutional power perpetuated by algorithmic decision-making, and driven by a push towards financial sustainability.
comment: In CHI '24 Proceedings of the CHI Conference on Human Factors in Computing Systems Honolulu, HI, USA
♻ ☆ LogPrécis: Unleashing Language Models for Automated Malicious Log Analysis
The collection of security-related logs holds the key to understanding attack behaviors and diagnosing vulnerabilities. Still, their analysis remains a daunting challenge. Recently, Language Models (LMs) have demonstrated unmatched potential in understanding natural and programming languages. The question arises whether and how LMs could be also useful for security experts since their logs contain intrinsically confused and obfuscated information. In this paper, we systematically study how to benefit from the state-of-the-art in LM to automatically analyze text-like Unix shell attack logs. We present a thorough design methodology that leads to LogPr\'ecis. It receives as input raw shell sessions and automatically identifies and assigns the attacker tactic to each portion of the session, i.e., unveiling the sequence of the attacker's goals. We demonstrate LogPr\'ecis capability to support the analysis of two large datasets containing about 400,000 unique Unix shell attacks. LogPr\'ecis reduces them into about 3,000 fingerprints, each grouping sessions with the same sequence of tactics. The abstraction it provides lets the analyst better understand attacks, identify fingerprints, detect novelty, link similar attacks, and track families and mutations. Overall, LogPr\'ecis, released as open source, paves the way for better and more responsive defense against cyberattacks.
comment: 18 pages, Computer&Security (https://www.sciencedirect.com/science/article/pii/S0167404824001068), code available at https://github.com/SmartData-Polito/logprecis, models available at https://huggingface.co/SmartDataPolito
♻ ☆ VisionGPT-3D: A Generalized Multimodal Agent for Enhanced 3D Vision Understanding
The evolution of text to visual components facilitates people's daily lives, such as generating image, videos from text and identifying the desired elements within the images. Computer vision models involving the multimodal abilities in the previous days are focused on image detection, classification based on well-defined objects. Large language models (LLMs) introduces the transformation from nature language to visual objects, which present the visual layout for text contexts. OpenAI GPT-4 has emerged as the pinnacle in LLMs, while the computer vision (CV) domain boasts a plethora of state-of-the-art (SOTA) models and algorithms to convert 2D images to their 3D representations. However, the mismatching between the algorithms with the problem could lead to undesired results. In response to this challenge, we propose an unified VisionGPT-3D framework to consolidate the state-of-the-art vision models, thereby facilitating the development of vision-oriented AI. VisionGPT-3D provides a versatile multimodal framework building upon the strengths of multimodal foundation models. It seamlessly integrates various SOTA vision models and brings the automation in the selection of SOTA vision models, identifies the suitable 3D mesh creation algorithms corresponding to 2D depth maps analysis, generates optimal results based on diverse multimodal inputs such as text prompts. Keywords: VisionGPT-3D, 3D vision understanding, Multimodal agent
comment: 12 pages, 7 figures, pending conference
♻ ☆ MSAC: Multiple Speech Attribute Control Method for Reliable Speech Emotion Recognition
Despite notable progress, speech emotion recognition (SER) remains challenging due to the intricate and ambiguous nature of speech emotion, particularly in wild world. While current studies primarily focus on recognition and generalization abilities, our research pioneers an investigation into the reliability of SER methods in the presence of semantic data shifts and explores how to exert fine-grained control over various attributes inherent in speech signals to enhance speech emotion modeling. In this paper, we first introduce MSAC-SERNet, a novel unified SER framework capable of simultaneously handling both single-corpus and cross-corpus SER. Specifically, concentrating exclusively on the speech emotion attribute, a novel CNN-based SER model is presented to extract discriminative emotional representations, guided by additive margin softmax loss. Considering information overlap between various speech attributes, we propose a novel learning paradigm based on correlations of different speech attributes, termed Multiple Speech Attribute Control (MSAC), which empowers the proposed SER model to simultaneously capture fine-grained emotion-related features while mitigating the negative impact of emotion-agnostic representations. Furthermore, we make a first attempt to examine the reliability of the MSAC-SERNet framework using out-of-distribution detection methods. Experiments on both single-corpus and cross-corpus SER scenarios indicate that MSAC-SERNet not only consistently outperforms the baseline in all aspects, but achieves superior performance compared to state-of-the-art SER approaches.
comment: 12 pages
♻ ☆ Task-Oriented GNNs Training on Large Knowledge Graphs for Accurate and Efficient Modeling ICDE
A Knowledge Graph (KG) is a heterogeneous graph encompassing a diverse range of node and edge types. Heterogeneous Graph Neural Networks (HGNNs) are popular for training machine learning tasks like node classification and link prediction on KGs. However, HGNN methods exhibit excessive complexity influenced by the KG's size, density, and the number of node and edge types. AI practitioners handcraft a subgraph of a KG G relevant to a specific task. We refer to this subgraph as a task-oriented subgraph (TOSG), which contains a subset of task-related node and edge types in G. Training the task using TOSG instead of G alleviates the excessive computation required for a large KG. Crafting the TOSG demands a deep understanding of the KG's structure and the task's objectives. Hence, it is challenging and time-consuming. This paper proposes KG-TOSA, an approach to automate the TOSG extraction for task-oriented HGNN training on a large KG. In KG-TOSA, we define a generic graph pattern that captures the KG's local and global structure relevant to a specific task. We explore different techniques to extract subgraphs matching our graph pattern: namely (i) two techniques sampling around targeted nodes using biased random walk or influence scores, and (ii) a SPARQL-based extraction method leveraging RDF engines' built-in indices. Hence, it achieves negligible preprocessing overhead compared to the sampling techniques. We develop a benchmark of real KGs of large sizes and various tasks for node classification and link prediction. Our experiments show that KG-TOSA helps state-of-the-art HGNN methods reduce training time and memory usage by up to 70% while improving the model performance, e.g., accuracy and inference time.
comment: 12 pages,9 Figures, 3 Tables, ICDE:2024
♻ ☆ Interpretability Guarantees with Merlin-Arthur Classifiers AISTATS24
We propose an interactive multi-agent classifier that provides provable interpretability guarantees even for complex agents such as neural networks. These guarantees consist of lower bounds on the mutual information between selected features and the classification decision. Our results are inspired by the Merlin-Arthur protocol from Interactive Proof Systems and express these bounds in terms of measurable metrics such as soundness and completeness. Compared to existing interactive setups, we rely neither on optimal agents nor on the assumption that features are distributed independently. Instead, we use the relative strength of the agents as well as the new concept of Asymmetric Feature Correlation which captures the precise kind of correlations that make interpretability guarantees difficult. We evaluate our results on two small-scale datasets where high mutual information can be verified explicitly.
comment: AISTATS24 Camera-Ready Version, 34 pages total (9 pages main part, 3 pages references, 22 pages appendix), 17 figures, 3 tables
♻ ☆ Can a GPT4-Powered AI Agent Be a Good Enough Performance Attribution Analyst?
Performance attribution analysis, defined as the process of explaining the drivers of the excess performance of an investment portfolio against a benchmark, stands as a significant feature of portfolio management and plays a crucial role in the investment decision-making process, particularly within the fund management industry. Rooted in a solid financial and mathematical framework, the importance and methodologies of this analytical technique are extensively documented across numerous academic research papers and books. The integration of large language models (LLMs) and AI agents marks a groundbreaking development in this field. These agents are designed to automate and enhance the performance attribution analysis by accurately calculating and analyzing portfolio performances against benchmarks. In this study, we introduce the application of an AI Agent for a variety of essential performance attribution tasks, including the analysis of performance drivers and utilizing LLMs as calculation engine for multi-level attribution analysis and question-answering (QA) tasks. Leveraging advanced prompt engineering techniques such as Chain-of-Thought (CoT) and Plan and Solve (PS), and employing a standard agent framework from LangChain, the research achieves promising results: it achieves accuracy rates exceeding 93% in analyzing performance drivers, attains 100% in multi-level attribution calculations, and surpasses 84% accuracy in QA exercises that simulate official examination standards. These findings affirm the impactful role of AI agents, prompt engineering and evaluation in advancing portfolio management processes, highlighting a significant development in the practical application and evaluation of Generative AI technologies within the domain.
♻ ☆ PIA: Your Personalized Image Animator via Plug-and-Play Modules in Text-to-Image Models
Recent advancements in personalized text-to-image (T2I) models have revolutionized content creation, empowering non-experts to generate stunning images with unique styles. While promising, adding realistic motions into these personalized images by text poses significant challenges in preserving distinct styles, high-fidelity details, and achieving motion controllability by text. In this paper, we present PIA, a Personalized Image Animator that excels in aligning with condition images, achieving motion controllability by text, and the compatibility with various personalized T2I models without specific tuning. To achieve these goals, PIA builds upon a base T2I model with well-trained temporal alignment layers, allowing for the seamless transformation of any personalized T2I model into an image animation model. A key component of PIA is the introduction of the condition module, which utilizes the condition frame and inter-frame affinity as input to transfer appearance information guided by the affinity hint for individual frame synthesis in the latent space. This design mitigates the challenges of appearance-related image alignment within and allows for a stronger focus on aligning with motion-related guidance.
comment: Project page: https://pi-animator.github.io/
♻ ☆ FunQA: Towards Surprising Video Comprehension
Surprising videos, such as funny clips, creative performances, or visual illusions, attract significant attention. Enjoyment of these videos is not simply a response to visual stimuli; rather, it hinges on the human capacity to understand (and appreciate) commonsense violations depicted in these videos. We introduce FunQA, a challenging video question-answering (QA) dataset specifically designed to evaluate and enhance the depth of video reasoning based on counter-intuitive and fun videos. Unlike most video QA benchmarks which focus on less surprising contexts, e.g., cooking or instructional videos, FunQA covers three previously unexplored types of surprising videos: 1) HumorQA, 2) CreativeQA, and 3) MagicQA. For each subset, we establish rigorous QA tasks designed to assess the model's capability in counter-intuitive timestamp localization, detailed video description, and reasoning around counter-intuitiveness. We also pose higher-level tasks, such as attributing a fitting and vivid title to the video and scoring the video creativity. In total, the FunQA benchmark consists of 312K free-text QA pairs derived from 4.3K video clips, spanning a total of 24 video hours. Moreover, we propose FunMentor, an agent designed for Vision-Language Models (VLMs) that uses multi-turn dialogues to enhance models' understanding of counter-intuitiveness. Extensive experiments with existing VLMs demonstrate the effectiveness of FunMentor and reveal significant performance gaps for the FunQA videos across spatial-temporal reasoning, visual-centered reasoning, and free-text generation.
comment: Project Page: https://funqa-benchmark.github.io/ Codebase: https://github.com/Jingkang50/FunQA
♻ ☆ Learning to Embed Time Series Patches Independently ICLR 2024
Masked time series modeling has recently gained much attention as a self-supervised representation learning strategy for time series. Inspired by masked image modeling in computer vision, recent works first patchify and partially mask out time series, and then train Transformers to capture the dependencies between patches by predicting masked patches from unmasked patches. However, we argue that capturing such patch dependencies might not be an optimal strategy for time series representation learning; rather, learning to embed patches independently results in better time series representations. Specifically, we propose to use 1) the simple patch reconstruction task, which autoencode each patch without looking at other patches, and 2) the simple patch-wise MLP that embeds each patch independently. In addition, we introduce complementary contrastive learning to hierarchically capture adjacent time series information efficiently. Our proposed method improves time series forecasting and classification performance compared to state-of-the-art Transformer-based models, while it is more efficient in terms of the number of parameters and training/inference time. Code is available at this repository: https://github.com/seunghan96/pits.
comment: ICLR 2024
♻ ☆ Soft Contrastive Learning for Time Series ICLR 2024
Contrastive learning has shown to be effective to learn representations from time series in a self-supervised way. However, contrasting similar time series instances or values from adjacent timestamps within a time series leads to ignore their inherent correlations, which results in deteriorating the quality of learned representations. To address this issue, we propose SoftCLT, a simple yet effective soft contrastive learning strategy for time series. This is achieved by introducing instance-wise and temporal contrastive loss with soft assignments ranging from zero to one. Specifically, we define soft assignments for 1) instance-wise contrastive loss by the distance between time series on the data space, and 2) temporal contrastive loss by the difference of timestamps. SoftCLT is a plug-and-play method for time series contrastive learning that improves the quality of learned representations without bells and whistles. In experiments, we demonstrate that SoftCLT consistently improves the performance in various downstream tasks including classification, semi-supervised learning, transfer learning, and anomaly detection, showing state-of-the-art performance. Code is available at this repository: https://github.com/seunghan96/softclt.
comment: ICLR 2024 Spotlight
CPA-Enhancer: Chain-of-Thought Prompted Adaptive Enhancer for Object Detection under Unknown Degradations
Object detection methods under known single degradations have been extensively investigated. However, existing approaches require prior knowledge of the degradation type and train a separate model for each, limiting their practical applications in unpredictable environments. To address this challenge, we propose a chain-of-thought (CoT) prompted adaptive enhancer, CPA-Enhancer, for object detection under unknown degradations. Specifically, CPA-Enhancer progressively adapts its enhancement strategy under the step-by-step guidance of CoT prompts, that encode degradation-related information. To the best of our knowledge, it's the first work that exploits CoT prompting for object detection tasks. Overall, CPA-Enhancer is a plug-and-play enhancement model that can be integrated into any generic detectors to achieve substantial gains on degraded images, without knowing the degradation type priorly. Experimental results demonstrate that CPA-Enhancer not only sets the new state of the art for object detection but also boosts the performance of other downstream vision tasks under unknown degradations.
♻ ☆ S2DM: Sector-Shaped Diffusion Models for Video Generation
Diffusion models have achieved great success in image generation. However, when leveraging this idea for video generation, we face significant challenges in maintaining the consistency and continuity across video frames. This is mainly caused by the lack of an effective framework to align frames of videos with desired temporal features while preserving consistent semantic and stochastic features. In this work, we propose a novel Sector-Shaped Diffusion Model (S2DM) whose sector-shaped diffusion region is formed by a set of ray-shaped reverse diffusion processes starting at the same noise point. S2DM can generate a group of intrinsically related data sharing the same semantic and stochastic features while varying on temporal features with appropriate guided conditions. We apply S2DM to video generation tasks, and explore the use of optical flow as temporal conditions. Our experimental results show that S2DM outperforms many existing methods in the task of video generation without any temporal-feature modelling modules. For text-to-video generation tasks where temporal conditions are not explicitly given, we propose a two-stage generation strategy which can decouple the generation of temporal features from semantic-content features. We show that, without additional training, our model integrated with another temporal conditions generative model can still achieve comparable performance with existing works. Our results can be viewd at https://s2dm.github.io/S2DM/.
comment: 17 pages, 6 figures
♻ ☆ Accurately Predicting Probabilities of Safety-Critical Rare Events for Intelligent Systems
Intelligent systems are increasingly integral to our daily lives, yet rare safety-critical events present significant latent threats to their practical deployment. Addressing this challenge hinges on accurately predicting the probability of safety-critical events occurring within a given time step from the current state, a metric we define as 'criticality'. The complexity of predicting criticality arises from the extreme data imbalance caused by rare events in high dimensional variables associated with the rare events, a challenge we refer to as the curse of rarity. Existing methods tend to be either overly conservative or prone to overlooking safety-critical events, thus struggling to achieve both high precision and recall rates, which severely limits their applicability. This study endeavors to develop a criticality prediction model that excels in both precision and recall rates for evaluating the criticality of safety-critical autonomous systems. We propose a multi-stage learning framework designed to progressively densify the dataset, mitigating the curse of rarity across stages. To validate our approach, we evaluate it in two cases: lunar lander and bipedal walker scenarios. The results demonstrate that our method surpasses traditional approaches, providing a more accurate and dependable assessment of criticality in intelligent systems.
♻ ☆ AI-Dentify: Deep learning for proximal caries detection on bitewing x-ray -- HUNT4 Oral Health Study
Background: Dental caries diagnosis requires the manual inspection of diagnostic bitewing images of the patient, followed by a visual inspection and probing of the identified dental pieces with potential lesions. Yet the use of artificial intelligence, and in particular deep-learning, has the potential to aid in the diagnosis by providing a quick and informative analysis of the bitewing images. Methods: A dataset of 13,887 bitewings from the HUNT4 Oral Health Study were annotated individually by six different experts, and used to train three different object detection deep-learning architectures: RetinaNet (ResNet50), YOLOv5 (M size), and EfficientDet (D0 and D1 sizes). A consensus dataset of 197 images, annotated jointly by the same six dentist, was used for evaluation. A five-fold cross validation scheme was used to evaluate the performance of the AI models. Results: he trained models show an increase in average precision and F1-score, and decrease of false negative rate, with respect to the dental clinicians. When compared against the dental clinicians, the YOLOv5 model shows the largest improvement, reporting 0.647 mean average precision, 0.548 mean F1-score, and 0.149 mean false negative rate. Whereas the best annotators on each of these metrics reported 0.299, 0.495, and 0.164 respectively. Conclusion: Deep-learning models have shown the potential to assist dental professionals in the diagnosis of caries. Yet, the task remains challenging due to the artifacts natural to the bitewing images.
comment: 24 pages, 5 figure, 7 tables
♻ ☆ Unimodal Multi-Task Fusion for Emotional Mimicry Prediction
In this study, we propose a methodology for the Emotional Mimicry Intensity (EMI) Estimation task within the context of the 6th Workshop and Competition on Affective Behavior Analysis in-the-wild. Our approach leverages the Wav2Vec 2.0 framework, pre-trained on a comprehensive podcast dataset, to extract a broad range of audio features encompassing both linguistic and paralinguistic elements. We enhance feature representation through a fusion technique that integrates individual features with a global mean vector, introducing global contextual insights into our analysis. Additionally, we incorporate a pre-trained valence-arousal-dominance (VAD) module from the Wav2Vec 2.0 model. Our fusion employs a Long Short-Term Memory (LSTM) architecture for efficient temporal analysis of audio data. Utilizing only the provided audio data, our approach demonstrates significant improvements over the established baseline.
♻ ☆ Similarity-based Label Inference Attack against Training and Inference of Split Learning
Split learning is a promising paradigm for privacy-preserving distributed learning. The learning model can be cut into multiple portions to be collaboratively trained at the participants by exchanging only the intermediate results at the cut layer. Understanding the security performance of split learning is critical for many privacy-sensitive applications. This paper shows that the exchanged intermediate results, including the smashed data (i.e., extracted features from the raw data) and gradients during training and inference of split learning, can already reveal the private labels. We mathematically analyze the potential label leakages and propose the cosine and Euclidean similarity measurements for gradients and smashed data, respectively. Then, the two similarity measurements are shown to be unified in Euclidean space. Based on the similarity metric, we design three label inference attacks to efficiently recover the private labels during both the training and inference phases. Experimental results validate that the proposed approaches can achieve close to 100% accuracy of label attacks. The proposed attack can still achieve accurate predictions against various state-of-the-art defense mechanisms, including DP-SGD, label differential privacy, gradient compression, and Marvell.
♻ ☆ Cross-domain Random Pre-training with Prototypes for Reinforcement Learning
This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. Unsupervised cross-domain Reinforcement Learning (RL) pre-training shows great potential for challenging continuous visual control but poses a big challenge. In this paper, we propose \textbf{C}ross-domain \textbf{R}andom \textbf{P}re-\textbf{T}raining with \textbf{pro}totypes (CRPTpro), a novel, efficient, and effective self-supervised cross-domain RL pre-training framework. CRPTpro decouples data sampling from encoder pre-training, proposing decoupled random collection to easily and quickly generate a qualified cross-domain pre-training dataset. Moreover, a novel prototypical self-supervised algorithm is proposed to pre-train an effective visual encoder that is generic across different domains. Without finetuning, the cross-domain encoder can be implemented for challenging downstream tasks defined in different domains, either seen or unseen. Compared with recent advanced methods, CRPTpro achieves better performance on downstream policy learning without extra training on exploration agents for data collection, greatly reducing the burden of pre-training. We conduct extensive experiments across eight challenging continuous visual-control domains, including balance control, robot locomotion, and manipulation. CRPTpro significantly outperforms the next best Proto-RL(C) on 11/12 cross-domain downstream tasks with only 54\% wall-clock pre-training time, exhibiting state-of-the-art pre-training performance with greatly improved pre-training efficiency. The complete code is available at https://github.com/liuxin0824/CRPTpro.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ E-Sparse: Boosting the Large Language Model Inference through Entropy-based N:M Sparsity
Traditional pruning methods are known to be challenging to work in Large Language Models (LLMs) for Generative AI because of their unaffordable training process and large computational demands. For the first time, we introduce the information entropy of hidden state features into a pruning metric design, namely E-Sparse, to improve the accuracy of N:M sparsity on LLM. E-Sparse employs the information richness to leverage the channel importance, and further incorporates several novel techniques to put it into effect: (1) it introduces information entropy to enhance the significance of parameter weights and input feature norms as a novel pruning metric, and performs N:M sparsity without modifying the remaining weights. (2) it designs global naive shuffle and local block shuffle to quickly optimize the information distribution and adequately cope with the impact of N:M sparsity on LLMs' accuracy. E-Sparse is implemented as a Sparse-GEMM on FasterTransformer and runs on NVIDIA Ampere GPUs. Extensive experiments on the LLaMA family and OPT models show that E-Sparse can significantly speed up the model inference over the dense model (up to 1.53X) and obtain significant memory saving (up to 43.52%), with acceptable accuracy loss.
♻ ☆ ToonAging: Face Re-Aging upon Artistic Portrait Style Transfer
Face re-aging is a prominent field in computer vision and graphics, with significant applications in photorealistic domains such as movies, advertising, and live streaming. Recently, the need to apply face re-aging to non-photorealistic images, like comics, illustrations, and animations, has emerged as an extension in various entertainment sectors. However, the lack of a network that can seamlessly edit the apparent age in NPR images has limited these tasks to a naive, sequential approach. This often results in unpleasant artifacts and a loss of facial attributes due to domain discrepancies. In this paper, we introduce a novel one-stage method for face re-aging combined with portrait style transfer, executed in a single generative step. We leverage existing face re-aging and style transfer networks, both trained within the same PR domain. Our method uniquely fuses distinct latent vectors, each responsible for managing aging-related attributes and NPR appearance. By adopting an exemplar-based approach, our method offers greater flexibility compared to domain-level fine-tuning approaches, which typically require separate training or fine-tuning for each domain. This effectively addresses the limitation of requiring paired datasets for re-aging and domain-level, data-driven approaches for stylization. Our experiments show that our model can effortlessly generate re-aged images while simultaneously transferring the style of examples, maintaining both natural appearance and controllability.
comment: 14 pages, 15 figures, 1 table
♻ ☆ Align-to-Distill: Trainable Attention Alignment for Knowledge Distillation in Neural Machine Translation LREC
The advent of scalable deep models and large datasets has improved the performance of Neural Machine Translation. Knowledge Distillation (KD) enhances efficiency by transferring knowledge from a teacher model to a more compact student model. However, KD approaches to Transformer architecture often rely on heuristics, particularly when deciding which teacher layers to distill from. In this paper, we introduce the 'Align-to-Distill' (A2D) strategy, designed to address the feature mapping problem by adaptively aligning student attention heads with their teacher counterparts during training. The Attention Alignment Module in A2D performs a dense head-by-head comparison between student and teacher attention heads across layers, turning the combinatorial mapping heuristics into a learning problem. Our experiments show the efficacy of A2D, demonstrating gains of up to +3.61 and +0.63 BLEU points for WMT-2022 De->Dsb and WMT-2014 En->De, respectively, compared to Transformer baselines.
comment: Accepted to LREC-COLING 2024
♻ ☆ A Benchmark Study on Calibration ICLR 2024
Deep neural networks are increasingly utilized in various machine learning tasks. However, as these models grow in complexity, they often face calibration issues, despite enhanced prediction accuracy. Many studies have endeavored to improve calibration performance through the use of specific loss functions, data preprocessing and training frameworks. Yet, investigations into calibration properties have been somewhat overlooked. Our study leverages the Neural Architecture Search (NAS) search space, offering an exhaustive model architecture space for thorough calibration properties exploration. We specifically create a model calibration dataset. This dataset evaluates 90 bin-based and 12 additional calibration measurements across 117,702 unique neural networks within the widely employed NATS-Bench search space. Our analysis aims to answer several longstanding questions in the field, using our proposed dataset: (i) Can model calibration be generalized across different datasets? (ii) Can robustness be used as a calibration measurement? (iii) How reliable are calibration metrics? (iv) Does a post-hoc calibration method affect all models uniformly? (v) How does calibration interact with accuracy? (vi) What is the impact of bin size on calibration measurement? (vii) Which architectural designs are beneficial for calibration? Additionally, our study bridges an existing gap by exploring calibration within NAS. By providing this dataset, we enable further research into NAS calibration. As far as we are aware, our research represents the first large-scale investigation into calibration properties and the premier study of calibration issues within NAS. The project page can be found at https://www.taolinwei.com/calibration-study
comment: ICLR 2024 poster
♻ ☆ Generalisable Agents for Neural Network Optimisation NeurIPS 2023
Optimising deep neural networks is a challenging task due to complex training dynamics, high computational requirements, and long training times. To address this difficulty, we propose the framework of Generalisable Agents for Neural Network Optimisation (GANNO) -- a multi-agent reinforcement learning (MARL) approach that learns to improve neural network optimisation by dynamically and responsively scheduling hyperparameters during training. GANNO utilises an agent per layer that observes localised network dynamics and accordingly takes actions to adjust these dynamics at a layerwise level to collectively improve global performance. In this paper, we use GANNO to control the layerwise learning rate and show that the framework can yield useful and responsive schedules that are competitive with handcrafted heuristics. Furthermore, GANNO is shown to perform robustly across a wide variety of unseen initial conditions, and can successfully generalise to harder problems than it was trained on. Our work presents an overview of the opportunities that this paradigm offers for training neural networks, along with key challenges that remain to be overcome.
comment: Accepted at the Workshop on Advanced Neural Network Training (WANT) and Optimization for Machine Learning (OPT) at NeurIPS 2023
♻ ☆ Zero-Shot Cross-Lingual Document-Level Event Causality Identification with Heterogeneous Graph Contrastive Transfer Learning LREC
Event Causality Identification (ECI) refers to the detection of causal relations between events in texts. However, most existing studies focus on sentence-level ECI with high-resource languages, leaving more challenging document-level ECI (DECI) with low-resource languages under-explored. In this paper, we propose a Heterogeneous Graph Interaction Model with Multi-granularity Contrastive Transfer Learning (GIMC) for zero-shot cross-lingual document-level ECI. Specifically, we introduce a heterogeneous graph interaction network to model the long-distance dependencies between events that are scattered over a document. Then, to improve cross-lingual transferability of causal knowledge learned from the source language, we propose a multi-granularity contrastive transfer learning module to align the causal representations across languages. Extensive experiments show our framework outperforms the previous state-of-the-art model by 9.4% and 8.2% of average F1 score on monolingual and multilingual scenarios respectively. Notably, in the multilingual scenario, our zero-shot framework even exceeds GPT-3.5 with few-shot learning by 24.3% in overall performance.
comment: Accepted at LREC-COLING 2024
♻ ☆ HealMe: Harnessing Cognitive Reframing in Large Language Models for Psychotherapy
Large Language Models (LLMs) can play a vital role in psychotherapy by adeptly handling the crucial task of cognitive reframing and overcoming challenges such as shame, distrust, therapist skill variability, and resource scarcity. Previous LLMs in cognitive reframing mainly converted negative emotions to positive ones, but these approaches have limited efficacy, often not promoting clients' self-discovery of alternative perspectives. In this paper, we unveil the Helping and Empowering through Adaptive Language in Mental Enhancement (HealMe) model. This novel cognitive reframing therapy method effectively addresses deep-rooted negative thoughts and fosters rational, balanced perspectives. Diverging from traditional LLM methods, HealMe employs empathetic dialogue based on psychotherapeutic frameworks. It systematically guides clients through distinguishing circumstances from feelings, brainstorming alternative viewpoints, and developing empathetic, actionable suggestions. Moreover, we adopt the first comprehensive and expertly crafted psychological evaluation metrics, specifically designed to rigorously assess the performance of cognitive reframing, in both AI-simulated dialogues and real-world therapeutic conversations. Experimental results show that our model outperforms others in terms of empathy, guidance, and logical coherence, demonstrating its effectiveness and potential positive impact on psychotherapy.
comment: 17 pages, 4 figures
♻ ☆ Promoting Segment Anything Model towards Highly Accurate Dichotomous Image Segmentation
The Segment Anything Model (SAM) represents a significant breakthrough into foundation models for computer vision, providing a large-scale image segmentation model. However, despite SAM's zero-shot performance, its segmentation masks lack fine-grained details, particularly in accurately delineating object boundaries. We have high expectations regarding whether SAM, as a foundation model, can be improved towards highly accurate object segmentation, which is known as dichotomous image segmentation (DIS). To address this issue, we propose DIS-SAM, which advances SAM towards DIS with extremely accurate details. DIS-SAM is a framework specifically tailored for highly accurate segmentation, maintaining SAM's promptable design. DIS-SAM employs a two-stage approach, integrating SAM with a modified IS-Net dedicated to DIS. Despite its simplicity, DIS-SAM demonstrates significantly enhanced segmentation accuracy compared to SAM and HQ-SAM.
♻ ☆ ID-like Prompt Learning for Few-Shot Out-of-Distribution Detection
Out-of-distribution (OOD) detection methods often exploit auxiliary outliers to train model identifying OOD samples, especially discovering challenging outliers from auxiliary outliers dataset to improve OOD detection. However, they may still face limitations in effectively distinguishing between the most challenging OOD samples that are much like in-distribution (ID) data, i.e., \idlike samples. To this end, we propose a novel OOD detection framework that discovers \idlike outliers using CLIP \cite{DBLP:conf/icml/RadfordKHRGASAM21} from the vicinity space of the ID samples, thus helping to identify these most challenging OOD samples. Then a prompt learning framework is proposed that utilizes the identified \idlike outliers to further leverage the capabilities of CLIP for OOD detection. Benefiting from the powerful CLIP, we only need a small number of ID samples to learn the prompts of the model without exposing other auxiliary outlier datasets. By focusing on the most challenging \idlike OOD samples and elegantly exploiting the capabilities of CLIP, our method achieves superior few-shot learning performance on various real-world image datasets (e.g., in 4-shot OOD detection on the ImageNet-1k dataset, our method reduces the average FPR95 by 12.16\% and improves the average AUROC by 2.76\%, compared to state-of-the-art methods). Code is available at https://github.com/ycfate/ID-like.
♻ ☆ KoCoSa: Korean Context-aware Sarcasm Detection Dataset
Sarcasm is a way of verbal irony where someone says the opposite of what they mean, often to ridicule a person, situation, or idea. It is often difficult to detect sarcasm in the dialogue since detecting sarcasm should reflect the context (i.e., dialogue history). In this paper, we introduce a new dataset for the Korean dialogue sarcasm detection task, KoCoSa (Korean Context-aware Sarcasm Detection Dataset), which consists of 12.8K daily Korean dialogues and the labels for this task on the last response. To build the dataset, we propose an efficient sarcasm detection dataset generation pipeline: 1) generating new sarcastic dialogues from source dialogues with large language models, 2) automatic and manual filtering of abnormal and toxic dialogues, and 3) human annotation for the sarcasm detection task. We also provide a simple but effective baseline for the Korean sarcasm detection task trained on our dataset. Experimental results on the dataset show that our baseline system outperforms strong baselines like large language models, such as GPT-3.5, in the Korean sarcasm detection task. We show that the sarcasm detection task relies deeply on the existence of sufficient context. We will release the dataset at https://github.com/Yu-billie/KoCoSa_sarcasm_detection.
♻ ☆ Predicting Generalization of AI Colonoscopy Models to Unseen Data
$\textbf{Background}$: Generalizability of AI colonoscopy algorithms is important for wider adoption in clinical practice. However, current techniques for evaluating performance on unseen data require expensive and time-intensive labels. $\textbf{Methods}$: We use a "Masked Siamese Network" (MSN) to identify novel phenomena in unseen data and predict polyp detector performance. MSN is trained to predict masked out regions of polyp images, without any labels. We test MSN's ability to be trained on data only from Israel and detect unseen techniques, narrow-band imaging (NBI) and chromendoscoy (CE), on colonoscopes from Japan (354 videos, 128 hours). We also test MSN's ability to predict performance of Computer Aided Detection (CADe) of polyps on colonoscopies from both countries, even though MSN is not trained on data from Japan. $\textbf{Results}$: MSN correctly identifies NBI and CE as less similar to Israel whitelight than Japan whitelight (bootstrapped z-test, |z| > 496, p < 10^-8 for both) using the label-free Frechet distance. MSN detects NBI with 99% accuracy, predicts CE better than our heuristic (90% vs 79% accuracy) despite being trained only on whitelight, and is the only method that is robust to noisy labels. MSN predicts CADe polyp detector performance on in-domain Israel and out-of-domain Japan colonoscopies (r=0.79, 0.37 respectively). With few examples of Japan detector performance to train on, MSN prediction of Japan performance improves (r=0.56). $\textbf{Conclusion}$: Our technique can identify distribution shifts in clinical data and can predict CADe detector performance on unseen data, without labels. Our self-supervised approach can aid in detecting when data in practice is different from training, such as between hospitals or data has meaningfully shifted from training. MSN has potential for application to medical image domains beyond colonoscopy.
♻ ☆ Decision-making with Speculative Opponent Models
Opponent modelling has proven effective in enhancing the decision-making of the controlled agent by constructing models of opponent agents. However, existing methods often rely on access to the observations and actions of opponents, a requirement that is infeasible when such information is either unobservable or challenging to obtain. To address this issue, we introduce Distributional Opponent-aided Multi-agent Actor-Critic (DOMAC), the first speculative opponent modelling algorithm that relies solely on local information (i.e., the controlled agent's observations, actions, and rewards). Specifically, the actor maintains a speculated belief about the opponents using the tailored speculative opponent models that predict the opponents' actions using only local information. Moreover, DOMAC features distributional critic models that estimate the return distribution of the actor's policy, yielding a more fine-grained assessment of the actor's quality. This thus more effectively guides the training of the speculative opponent models that the actor depends upon. Furthermore, we formally derive a policy gradient theorem with the proposed opponent models. Extensive experiments under eight different challenging multi-agent benchmark tasks within the MPE, Pommerman and StarCraft Multiagent Challenge (SMAC) demonstrate that our DOMAC successfully models opponents' behaviours and delivers superior performance against state-of-the-art methods with a faster convergence speed.
comment: 13 pages, 27 figures
♻ ☆ On Image Search in Histopathology
Pathology images of histopathology can be acquired from camera-mounted microscopes or whole slide scanners. Utilizing similarity calculations to match patients based on these images holds significant potential in research and clinical contexts. Recent advancements in search technologies allow for implicit quantification of tissue morphology across diverse primary sites, facilitating comparisons and enabling inferences about diagnosis, and potentially prognosis, and predictions for new patients when compared against a curated database of diagnosed and treated cases. In this paper, we comprehensively review the latest developments in image search technologies for histopathology, offering a concise overview tailored for computational pathology researchers seeking effective, fast and efficient image search methods in their work.
comment: A chapter in the Book "Artificial INtelligence in Digital Pathology" by Cohen and Chauhan, 2024
♻ ☆ Physics-Enhanced Multi-fidelity Learning for Optical Surface Imprint
Human fingerprints serve as one unique and powerful characteristic for each person, from which policemen can recognize the identity. Similar to humans, many natural bodies and intrinsic mechanical qualities can also be uniquely identified from surface characteristics. To measure the elasto-plastic properties of one material, one formally sharp indenter is pushed into the measured body under constant force and retracted, leaving a unique residual imprint of the minute size from several micrometers to nanometers. However, one great challenge is how to map the optical image of this residual imprint into the real wanted mechanical properties, \ie, the tensile force curve. In this paper, we propose a novel method to use multi-fidelity neural networks (MFNN) to solve this inverse problem. We first build up the NN model via pure simulation data, and then bridge the sim-to-real gap via transfer learning. Considering the difficulty of collecting real experimental data, we use NN to dig out the unknown physics and also implant the known physics into the transfer learning framework, thus highly improving the model stability and decreasing the data requirement. The final constructed model only needs three-shot calibration of real materials. We tested the final model across 20 real materials and achieved satisfying accuracy. This work serves as one great example of applying machine learning into scientific research, especially under the constraints of data limitation and fidelity variance.
comment: 15 pages, 11 figure
♻ ☆ Hulk: A Universal Knowledge Translator for Human-Centric Tasks
Human-centric perception tasks, e.g., pedestrian detection, skeleton-based action recognition, and pose estimation, have wide industrial applications, such as metaverse and sports analysis. There is a recent surge to develop human-centric foundation models that can benefit a broad range of human-centric perception tasks. While many human-centric foundation models have achieved success, they did not explore 3D and vision-language tasks for human-centric and required task-specific finetuning. These limitations restrict their application to more downstream tasks and situations. To tackle these problems, we present Hulk, the first multimodal human-centric generalist model, capable of addressing 2D vision, 3D vision, skeleton-based, and vision-language tasks without task-specific finetuning. The key to achieving this is condensing various task-specific heads into two general heads, one for discrete representations, e.g., languages, and the other for continuous representations, e.g., location coordinates. The outputs of two heads can be further stacked into four distinct input and output modalities. This uniform representation enables Hulk to treat diverse human-centric tasks as modality translation, integrating knowledge across a wide range of tasks. Comprehensive evaluations of Hulk on 12 benchmarks covering 8 human-centric tasks demonstrate the superiority of our proposed method, achieving state-of-the-art performance in 11 benchmarks. The code is available on https://github.com/OpenGVLab/Hulk.
comment: 24 pages, 5 figures
♻ ☆ AnyV2V: A Plug-and-Play Framework For Any Video-to-Video Editing Tasks
Video-to-video editing involves editing a source video along with additional control (such as text prompts, subjects, or styles) to generate a new video that aligns with the source video and the provided control. Traditional methods have been constrained to certain editing types, limiting their ability to meet the wide range of user demands. In this paper, we introduce AnyV2V, a novel training-free framework designed to simplify video editing into two primary steps: (1) employing an off-the-shelf image editing model (e.g. InstructPix2Pix, InstantID, etc) to modify the first frame, (2) utilizing an existing image-to-video generation model (e.g. I2VGen-XL) for DDIM inversion and feature injection. In the first stage, AnyV2V can plug in any existing image editing tools to support an extensive array of video editing tasks. Beyond the traditional prompt-based editing methods, AnyV2V also can support novel video editing tasks, including reference-based style transfer, subject-driven editing, and identity manipulation, which were unattainable by previous methods. In the second stage, AnyV2V can plug in any existing image-to-video models to perform DDIM inversion and intermediate feature injection to maintain the appearance and motion consistency with the source video. On the prompt-based editing, we show that AnyV2V can outperform the previous best approach by 35\% on prompt alignment, and 25\% on human preference. On the three novel tasks, we show that AnyV2V also achieves a high success rate. We believe AnyV2V will continue to thrive due to its ability to seamlessly integrate the fast-evolving image editing methods. Such compatibility can help AnyV2V to increase its versatility to cater to diverse user demands.
comment: preprint
♻ ☆ Enhancing Worker Recruitment in Collaborative Mobile Crowdsourcing: A Graph Neural Network Trust Evaluation Approach
Collaborative Mobile Crowdsourcing (CMCS) allows platforms to recruit worker teams to collaboratively execute complex sensing tasks. The efficiency of such collaborations could be influenced by trust relationships among workers. To obtain the asymmetric trust values among all workers in the social network, the Trust Reinforcement Evaluation Framework (TREF) based on Graph Convolutional Neural Networks (GCNs) is proposed in this paper. The task completion effect is comprehensively calculated by considering the workers' ability benefits, distance benefits, and trust benefits in this paper. The worker recruitment problem is modeled as an Undirected Complete Recruitment Graph (UCRG), for which a specific Tabu Search Recruitment (TSR) algorithm solution is proposed. An optimal execution team is recruited for each task by the TSR algorithm, and the collaboration team for the task is obtained under the constraint of privacy loss. To enhance the efficiency of the recruitment algorithm on a large scale and scope, the Mini-Batch K-Means clustering algorithm and edge computing technology are introduced, enabling distributed worker recruitment. Lastly, extensive experiments conducted on five real datasets validate that the recruitment algorithm proposed in this paper outperforms other baselines. Additionally, TREF proposed herein surpasses the performance of state-of-the-art trust evaluation methods in the literature.
comment: The article has been accepted by IEEE TMC, and its DOI is 10.1109/TMC.2024.3373469
♻ ☆ Scalable Optimal Transport Methods in Machine Learning: A Contemporary Survey
Optimal Transport (OT) is a mathematical framework that first emerged in the eighteenth century and has led to a plethora of methods for answering many theoretical and applied questions. The last decade has been a witness to the remarkable contributions of this classical optimization problem to machine learning. This paper is about where and how optimal transport is used in machine learning with a focus on the question of scalable optimal transport. We provide a comprehensive survey of optimal transport while ensuring an accessible presentation as permitted by the nature of the topic and the context. First, we explain the optimal transport background and introduce different flavors (i.e., mathematical formulations), properties, and notable applications. We then address the fundamental question of how to scale optimal transport to cope with the current demands of big and high dimensional data. We conduct a systematic analysis of the methods used in the literature for scaling OT and present the findings in a unified taxonomy. We conclude with presenting some open challenges and discussing potential future research directions. A live repository of related OT research papers is maintained in https://github.com/abdelwahed/OT_for_big_data.git
comment: Accepted @ TPAMI 24
♻ ☆ Evaluating Large Language Models as Generative User Simulators for Conversational Recommendation NAACL 2024
Synthetic users are cost-effective proxies for real users in the evaluation of conversational recommender systems. Large language models show promise in simulating human-like behavior, raising the question of their ability to represent a diverse population of users. We introduce a new protocol to measure the degree to which language models can accurately emulate human behavior in conversational recommendation. This protocol is comprised of five tasks, each designed to evaluate a key property that a synthetic user should exhibit: choosing which items to talk about, expressing binary preferences, expressing open-ended preferences, requesting recommendations, and giving feedback. Through evaluation of baseline simulators, we demonstrate these tasks effectively reveal deviations of language models from human behavior, and offer insights on how to reduce the deviations with model selection and prompting strategies.
comment: NAACL 2024
♻ ☆ A Unified Model for Longitudinal Multi-Modal Multi-View Prediction with Missingness
Medical records often consist of different modalities, such as images, text, and tabular information. Integrating all modalities offers a holistic view of a patient's condition, while analyzing them longitudinally provides a better understanding of disease progression. However, real-world longitudinal medical records present challenges: 1) patients may lack some or all of the data for a specific timepoint, and 2) certain modalities or views might be absent for all patients during a particular period. In this work, we introduce a unified model for longitudinal multi-modal multi-view prediction with missingness. Our method allows as many timepoints as desired for input, and aims to leverage all available data, regardless of their availability. We conduct extensive experiments on the knee osteoarthritis dataset from the Osteoarthritis Initiative for pain and Kellgren-Lawrence grade prediction at a future timepoint. We demonstrate the effectiveness of our method by comparing results from our unified model to specific models that use the same modality and view combinations during training and evaluation. We also show the benefit of having extended temporal data and provide post-hoc analysis for a deeper understanding of each modality/view's importance for different tasks.
Optimization and Control 42
Optimal Exploration Strategy for Regret Minimization in Unconstrained Scalar Optimization Problems IEEE Conference on Decision and Control
We study the problem of determining the optimal exploration strategy in an unconstrained scalar optimization problem depending on an unknown parameter to be learned from online collected noisy data. An optimal trade-off between exploration and exploitation is crucial for effective optimization under uncertainties, and to achieve this we consider a cumulative regret minimization approach over a finite horizon, with each time instant in the horizon characterized by a stochastic exploration signal, whose variance has to be designed. In this setting, under an idealized assumption on an appropriately defined information function associated with the excitation, we are able to show that the optimal exploration strategy is either to use no exploration at all (called lazy exploration) or adding an exploration excitation only at the first time instant of the horizon (called immediate exploration). A quadratic numerical example is used to illustrate the results.
comment: Preprint submitted to IEEE Conference on Decision and Control (CDC 2024)
☆ Quantitative propagation of smallness and spectral estimates for the Schrödinger operator
In this paper, we investigate quantitative propagation of smallness properties for the Schr\"odinger operator on a bounded domain in $\mathbb R^d$. We extend Logunov, Malinnikova's results concerning propagation of smallness for $A$-harmonic functions to solutions of divergence elliptic equations perturbed by a bounded zero order term. We also prove similar results for gradient of solutions to some particular equations. This latter result enables us to follow the recent strategy of Burq, Moyano for the obtaining of spectral estimates on rough sets for the Schr\"odinger operator. Applications to observability estimates and to the null-controllability of associated parabolic equations posed on compact manifolds or the whole euclidean space are then considered.
comment: Comments welcome
☆ Nonlinear Reachable Set Computation and Model Predictive Control for Safe Hypersonic Re-entry of Atmospheric Vehicles
This paper investigates the application of reachability analysis to the re-entry problem faced by vehicles entering Earth's atmosphere. The study delves into the time evolution of reachable sets for the system, particularly when subject to nonlinear implicit controls, given the potential damage from the intense heat generated during hypersonic re-entry. Our proposed methodology leverages zonotopes and constrained zonotopes to ensure compliance with safety specifications. Furthermore, we utilize Model Predictive Control for detailed trajectory planning. To substantiate our methodology, we provide detailed simulations that not only tackle nonlinear re-entry scenarios but also illustrate trajectory planning using MPC.
comment: 2023 AAS/AIAA Astrodynamics Specialist Conference, Big Sky, MT
☆ A data-driven approach to PDE-constrained optimization in inverse problems
Inverse problems are ubiquitous in science and engineering. Many of these are naturally formulated as a PDE-constrained optimization problem. These non-linear, large-scale, constrained optimization problems know many challenges, of which the inherent non-linearity of the problem is an important one. As an alternative to this physics-driven approach, data-driven methods have been proposed. These methods come with their own set of challenges, and it appears that, ideally, one would devise hybrid methods that combine the best of both worlds. In this paper, we propose one way of combining PDE-constrained optimization with recently proposed data-driven reduced-order models. Starting from an infinite-dimensional formulation of the inverse problem with discrete data, we propose a general framework for the analysis and discretisation of such problems. The proposed approach is based on a relaxed formulation of the PDE-constrained optimization problem, which reduces to a weighted non-linear least-squares problem. The weight matrix turns out to be the Gram matrix of solutions of the PDE, and it can be estimated directly from the measurements. We provide a number of representative case studies and numerical examples.
☆ A Stochastic Quasi-Newton Method for Non-convex Optimization with Non-uniform Smoothness
Classical convergence analyses for optimization algorithms rely on the widely-adopted uniform smoothness assumption. However, recent experimental studies have demonstrated that many machine learning problems exhibit non-uniform smoothness, meaning the smoothness factor is a function of the model parameter instead of a universal constant. In particular, it has been observed that the smoothness grows with respect to the gradient norm along the training trajectory. Motivated by this phenomenon, the recently introduced $(L_0, L_1)$-smoothness is a more general notion, compared to traditional $L$-smoothness, that captures such positive relationship between smoothness and gradient norm. Under this type of non-uniform smoothness, existing literature has designed stochastic first-order algorithms by utilizing gradient clipping techniques to obtain the optimal $\mathcal{O}(\epsilon^{-3})$ sample complexity for finding an $\epsilon$-approximate first-order stationary solution. Nevertheless, the studies of quasi-Newton methods are still lacking. Considering higher accuracy and more robustness for quasi-Newton methods, in this paper we propose a fast stochastic quasi-Newton method when there exists non-uniformity in smoothness. Leveraging gradient clipping and variance reduction, our algorithm can achieve the best-known $\mathcal{O}(\epsilon^{-3})$ sample complexity and enjoys convergence speedup with simple hyperparameter tuning. Our numerical experiments show that our proposed algorithm outperforms the state-of-the-art approaches.
☆ On moment relaxations for linear state feedback controller synthesis with non-convex quadratic costs and constraints IEEE Conference on Decision and Control
We present a simple and effective way to account for non-convex costs and constraints~in~state feedback synthesis, and an interpretation for the variables in which state feedback synthesis is typically convex. We achieve this by deriving the controller design using moment matrices of state and input. It turns out that this approach allows the consideration of non-convex constraints by relaxing them as expectation constraints, and that the variables in which state feedback synthesis is typically convexified can be identified with blocks of these moment matrices.
comment: Preprent to be submitted to IEEE Conference on Decision and Control
☆ Robust Microgrid Dispatch with Real-Time Energy Sharing and Endogenous Uncertainty
With the rising adoption of distributed energy resources (DERs), microgrid dispatch is facing new challenges: DER owners are independent stakeholders seeking to maximize their individual profits rather than being controlled centrally; and the dispatch of renewable generators may affect the microgrid's exposure to uncertainty. To address these challenges, this paper proposes a two-stage robust microgrid dispatch model with real-time energy sharing and endogenous uncertainty. In the day-ahead stage, the connection/disconnection of renewable generators is optimized, which influences the size and dimension of the uncertainty set. As a result, the uncertainty set is endogenously given. In addition, non-anticipative operational bounds for energy storage (ES) are derived to enable the online operation of ES in real-time. In the real-time stage, DER owners (consumers and prosumers) share energy with each other via a proposed energy sharing mechanism, which forms a generalized Nash game. To solve the robust microgrid dispatch model, we develop an equivalent optimization model to compute the real-time energy sharing equilibrium. Based on this, a projection-based column-and-constraint generation (C&CG) method is proposed to handle the endogenous uncertainty. Numerical experiments show the effectiveness and advantages of the proposed model and method.
comment: 11 pages, 11 figures
☆ Pursuit-Evasion on a Sphere and When It Can Be Considered Flat
In classical works on a planar differential pursuit-evasion game with a faster pursuer, the intercept point resulting from the equilibrium strategies lies on the Apollonius circle. This property was exploited for the construction of the equilibrium strategies for two faster pursuers against one evader. Extensions for planar multiple-pursuer single-evader scenarios have been considered. We study a pursuit-evasion game on a sphere and the relation of the equilibrium intercept point to the Apollonius domain on the sphere. The domain is a generalization of the planar Apollonius circle set. We find a condition resulting in the intercept point belonging to the Apollonius domain, which is the characteristic of the planar game solution. Finally, we use this characteristic to discuss pursuit and evasion strategies in the context of two pursuers and a single slower evader on the sphere and illustrate it using numerical simulations.
comment: 8 Pages, 5 figures, To be submitted to 2024 Conference on Decision and Control in Milan, Italy
☆ Near-optimal performance of stochastic economic MPC
This paper presents first results for near optimality in expectation of the closed-loop solutions for stochastic economic MPC. The approach relies on a recently developed turnpike property for stochastic optimal control problems at an optimal stationary process, combined with techniques for analyzing time-varying economic MPC schemes. We obtain near optimality in finite time as well as overtaking and average near optimality on infinite time horizons.
☆ On the Convergence of Adam under Non-uniform Smoothness: Separability from SGDM and Beyond
This paper aims to clearly distinguish between Stochastic Gradient Descent with Momentum (SGDM) and Adam in terms of their convergence rates. We demonstrate that Adam achieves a faster convergence compared to SGDM under the condition of non-uniformly bounded smoothness. Our findings reveal that: (1) in deterministic environments, Adam can attain the known lower bound for the convergence rate of deterministic first-order optimizers, whereas the convergence rate of Gradient Descent with Momentum (GDM) has higher order dependence on the initial function value; (2) in stochastic setting, Adam's convergence rate upper bound matches the lower bounds of stochastic first-order optimizers, considering both the initial function value and the final error, whereas there are instances where SGDM fails to converge with any learning rate. These insights distinctly differentiate Adam and SGDM regarding their convergence rates. Additionally, by introducing a novel stopping-time based technique, we further prove that if we consider the minimum gradient norm during iterations, the corresponding convergence rate can match the lower bounds across all problem hyperparameters. The technique can also help proving that Adam with a specific hyperparameter scheduler is parameter-agnostic, which hence can be of independent interest.
☆ Hybrid integrator-gain system based integral resonant controllers for negative imaginary systems
We introduce a hybrid control system called a hybrid integrator-gain system (HIGS) based integral resonant controller (IRC) to stabilize negative imaginary (NI) systems. A HIGS-based IRC has a similar structure to an IRC, with the integrator replaced by a HIGS. We show that a HIGS-based IRC is an NI system. Also, for a SISO NI system with a minimal realization, we show there exists a HIGS-based IRC such that their closed-loop interconnection is asymptotically stable. Also, we propose a proportional-integral-double-integral resonant controller and a HIGS-based proportional-integral-double-integral resonant controller and show that both of them can be applied to asymptotically stabilize an NI system. An example is provided to illustrate the proposed results.
comment: 9 pages, 9 figures
☆ Paddy: Evolutionary Optimization Algorithm for Chemical Systems and Spaces
Optimization of chemical systems and processes have been enhanced and enabled by the guidance of algorithms and analytical approaches. While many methods will systematically investigate how underlying variables govern a given outcome, there is often a substantial number of experiments needed to accurately model these relations. As chemical systems increase in complexity, inexhaustive processes must propose experiments that efficiently optimize the underlying objective, while ideally avoiding convergence on unsatisfactory local minima. We have developed the Paddy software package around the Paddy Field Algorithm, a biologically inspired evolutionary optimization algorithm that propagates parameters without direct inference of the underlying objective function. Benchmarked against the Tree of Parzen Estimator, a Bayesian algorithm implemented in the Hyperopt software Library, Paddy displays efficient optimization with lower runtime, and avoidance of early convergence. Herein we report these findings for the cases of: global optimization of a two-dimensional bimodal distribution, interpolation of an irregular sinusoidal function, hyperparameter optimization of an artificial neural network tasked with classification of solvent for reaction components, and targeted molecule generation via optimization of input vectors for a decoder network. We anticipate that the facile nature of Paddy will serve to aid in automated experimentation, where minimization of investigative trials and or diversity of suitable solutions is of high priority.
Optimal Contract Design for End-of-Life Care Payments
A large fraction of total healthcare expenditure occurs due to end-of-life (EOL) care, which means it is important to study the problem of more carefully incentivizing necessary versus unnecessary EOL care because this has the potential to reduce overall healthcare spending. This paper introduces a principal-agent model that integrates a mixed payment system of fee-for-service and pay-for-performance in order to analyze whether it is possible to better align healthcare provider incentives with patient outcomes and cost-efficiency in EOL care. The primary contributions are to derive optimal contracts for EOL care payments using a principal-agent framework under three separate models for the healthcare provider, where each model considers a different level of risk tolerance for the provider. We derive these optimal contracts by converting the underlying principal-agent models from a bilevel optimization problem into a single-level optimization problem that can be analytically solved. Our results are demonstrated using a simulation where an optimal contract is used to price intracranial pressure monitoring for traumatic brain injuries.
☆ Perturbations in PDE-constrained optimal control decay exponentially in space
For linear-quadratic optimal control problems (OCPs) governed by elliptic and parabolic partial differential equations (PDEs), we investigate the impact of perturbations on optimal solutions. Local perturbations may occur, e.g., due to discretization of the optimality system or disturbed problem data. Whereas these perturbations may exhibit global effects in the uncontrolled case, we prove that the ramifications are exponentially damped in space under stabilizability- and detectability-like conditions. To this end, we prove a bound on the optimality condition's solution operator that is uniform in the domain size. Then, this uniformity is used in a scaling argument to show the exponential decay of perturbations in space. We numerically validate and illustrate our results by solving OCPs involving Helmholtz, Poisson, and advection-diffusion-reaction equations.
comment: 26 pages, 9 figures
Optimal control of gradient flows via the Weighted Energy-Dissipation method
We consider a general optimal control problem in the setting of gradient flows. Two approximations of the problem are presented, both relying on the variational reformulation of gradient-flow dynamics via the Weighted-Energy-Dissipation variational approach. This consists in the minimization of global-in-time functionals over trajectories, combined with a limit passage. We show that the original nonpenalized problem and the two successive approximations admits solutions. Moreover, resorting to a $\Gamma$-convergence analysis we show that penalised optimal controls converge to nonpenalized one as the approximation is removed.
comment: 20 pages
☆ Riemannian Optimization and the Hartree-Fock Method
In the present work we studied a subfield of Applied Mathematics called Riemannian Optimization. The main goal of this subfield is to generalize algorithms, theorems and tools from Mathematical Optimization to the case in which the optimization problem is defined on a Riemannian manifold. As a case study, we implemented some of the main algorithms described in the literature (Gradient Descent, Newton-Raphson and Conjugate Gradient) to solve an optimization problem known as Hartree-Fock. This method is extremely important in the field of Computational Quantum Chemistry and it is a good case study because it is a problem somewhat hard to solve and, as a consequence of this, it requires many tools from Riemannian Optimization. Besides, it is also a good example to see how these algorithms perform in practice.
☆ Extragradient Sliding for Composite Non-Monotone Variational Inequalities
Variational inequalities offer a versatile and straightforward approach to analyzing a broad range of equilibrium problems in both theoretical and practical fields. In this paper, we consider a composite generally non-monotone variational inequality represented as a sum of $L_q$-Lipschitz monotone and $L_p$-Lipschitz generally non-monotone operators. We applied a special sliding version of the classical Extragradient method to this problem and obtain better convergence results. In particular, to achieve $\varepsilon$-accuracy of the solution, the oracle complexity of the non-monotone operator $Q$ for our algorithm is $O\left(L_p^2/\varepsilon^2\right)$ in contrast to the basic Extragradient algorithm with $O\left((L_p+L_q)^2/\varepsilon^2\right)$. The results of numerical experiments confirm the theoretical findings and show the superiority of the proposed method.
comment: 12 pages, 1 algorithm, 3 figures
☆ Model Construction for Convex-Constrained Derivative-Free Optimization
We develop a new approximation theory for linear and quadratic interpolation models, suitable for use in convex-constrained derivative-free optimization (DFO). Most existing model-based DFO methods for constrained problems assume the ability to construct sufficiently accurate approximations via interpolation, but the standard notions of accuracy (designed for unconstrained problems) may not be achievable by only sampling feasible points, and so may not give practical algorithms. This work extends the theory of convex-constrained linear interpolation developed in [Hough & Roberts, SIAM J. Optim, 32:4 (2022), pp. 2552-2579] to the case of linear regression models and underdetermined quadratic interpolation models.
☆ Adapprox: Adaptive Approximation in Adam Optimization via Randomized Low-Rank Matrices
As deep learning models exponentially increase in size, optimizers such as Adam encounter significant memory consumption challenges due to the storage of first and second moment data. Current memory-efficient methods like Adafactor and CAME often compromise accuracy with their matrix factorization techniques. Addressing this, we introduce Adapprox, a novel approach that employs randomized low-rank matrix approximation for a more effective and accurate approximation of Adam's second moment. Adapprox features an adaptive rank selection mechanism, finely balancing accuracy and memory efficiency, and includes an optional cosine similarity guidance strategy to enhance stability and expedite convergence. In GPT-2 training and downstream tasks, Adapprox surpasses AdamW by achieving 34.5% to 49.9% and 33.8% to 49.9% memory savings for the 117M and 345M models, respectively, with the first moment enabled, and further increases these savings without the first moment. Besides, it enhances convergence speed and improves downstream task performance relative to its counterparts.
☆ Data-Driven Predictive Control with Adaptive Disturbance Attenuation for Constrained Systems
In this paper, we propose a novel data-driven predictive control approach for systems subject to time-domain constraints. The approach combines the strengths of H-infinity control for rejecting disturbances and MPC for handling constraints. In particular, the approach can dynamically adapt H-infinity disturbance attenuation performance depending on measured system state and forecasted disturbance level to satisfy constraints. We establish theoretical properties of the approach including robust guarantees of closed-loop stability, disturbance attenuation, constraint satisfaction under noisy data, as well as sufficient conditions for recursive feasibility, and illustrate the approach with a numerical example.
comment: 11 pages, 2 figures
☆ A Stochastic Model-Based Control Methodology for Glycemic Management in the Intensive Care Unit
Intensive care unit (ICU) patients exhibit erratic blood glucose (BG) fluctuations, including hypoglycemic and hyperglycemic episodes, and require exogenous insulin delivery to keep their BG in healthy ranges. Glycemic control via glycemic management (GM) is associated with reduced mortality and morbidity in the ICU, but GM increases the cognitive load on clinicians. The availability of robust, accurate, and actionable clinical decision support (CDS) tools reduces this burden and assists in the decision-making process to improve health outcomes. Clinicians currently follow GM protocol flow charts for patient intravenous insulin delivery rate computations. We present a mechanistic model-based control algorithm that predicts the optimal intravenous insulin rate to keep BG within a target range; the goal is to develop this approach for eventual use within CDS systems. In this control framework, we employed a stochastic model representing BG dynamics in the ICU setting and used the linear quadratic Gaussian control methodology to develop a controller. We designed two experiments, one using virtual (simulated) patients and one using a real-world retrospective dataset. Using these, we evaluate the safety and efficacy of this model-based glycemic control methodology. The presented controller avoids hypoglycemia and hyperglycemia in virtual patients, maintaining BG levels in the target range more consistently than two existing GM protocols. Moreover, this methodology could theoretically prevent a large proportion of hypoglycemic and hyperglycemic events recorded in a real-world retrospective dataset.
comment: 14 pages, 4 figures, 5 tables
☆ Anderson acceleration of derivative-free projection methods for constrained monotone nonlinear equations
The derivative-free projection method (DFPM) is an efficient algorithm for solving monotone nonlinear equations. As problems grow larger, there is a strong demand for speeding up the convergence of DFPM. This paper considers the application of Anderson acceleration (AA) to DFPM for constrained monotone nonlinear equations. By employing a nonstationary relaxation parameter and interleaving with slight modifications in each iteration, a globally convergent variant of AA for DFPM named as AA-DFPM is proposed. Further, the linear convergence rate is proved under some mild assumptions. Experiments on both mathematical examples and a real-world application show encouraging results of AA-DFPM and confirm the suitability of AA for accelerating DFPM in solving optimization problems.
☆ Network Learning with Directional Sign Patterns
Complex systems can be effectively modeled via graphs that encode networked interactions, where relations between entities or nodes are often quantified by signed edge weights, e.g., promotion/inhibition in gene regulatory networks, or encoding political of friendship differences in social networks. However, it is often the case that only an aggregate consequence of such edge weights that characterize relations may be directly observable, as in protein expression of in gene regulatory networks. Thus, learning edge weights poses a significant challenge that is further exacerbated for intricate and large-scale networks. In this article, we address a model problem to determine the strength of sign-indefinite relations that explain marginal distributions that constitute our data. To this end, we develop a paradigm akin to that of the Schr\"odinger bridge problem and an efficient Sinkhorn type algorithm (more properly, Schr\"odinger-Fortet-Sinkhorn algorithm) that allows fast convergence to parameters that minimize a relative entropy/likelihood criterion between the sought signed adjacency matrix and a prior. The formalism that we present represents a novel generalization of the earlier Schr\"odinger formalism in that marginal computations may incorporate weights that model directionality in underlying relations, and further, that it can be extended to high-order networks -- the Schr\"odinger-Fortet-Sinkhorn algorithm that we derive is applicable all the same and allows geometric convergence to a sought sign-indefinite adjacency matrix or tensor, for high-order networks. We demonstrate our framework with synthetic and real-world examples.
☆ Fourier Galerkin approximation of mean field control problems
The purpose of this work is to provide a finite dimensional approximation of the solution to a mean field optimal control problem set on the $d$-dimensional torus. The approximation is obtained by means of a Fourier-Galerkin method, the main principle of which is to convolve probability measures on the torus by the Dirichlet kernel or, equivalently, to truncate the Fourier expansion of probability measures on the torus. However, this operation has the main feature not to leave the space of probability measures invariant, which drawback is know as \textit{Gibbs}' phenomenon. In spite of this, we manage to prove that, for initial conditions in the `interior' of the space of probability measures and for sufficiently large levels of truncation, the Fourier-Galerkin method induces a new finite dimensional control problem whose trajectories take values in the space of probability measures with a finite number of Fourier coefficients. Our main result asserts that, whenever the cost functionals are smooth and convex, the distance between the optimal trajectories of the original and approximating control problems decreases at a polynomial rate as the index of truncation in the Fourier-Galerkin method tends to $\infty$. A similar result holds for the distance between the corresponding value functions. From a practical point of view, our approach provides an efficient strategy to approximate mean field control optimizers by finite dimensional parameters and opens new perspectives for the numerical analysis of mean field control problems. It may be also applied to discretize more general mean field game systems.
☆ Data-Driven Approximation of Stationary Nonlinear Filters with Optimal Transport Maps
The nonlinear filtering problem is concerned with finding the conditional probability distribution (posterior) of the state of a stochastic dynamical system, given a history of partial and noisy observations. This paper presents a data-driven nonlinear filtering algorithm for the case when the state and observation processes are stationary. The posterior is approximated as the push-forward of an optimal transport (OT) map from a given distribution, that is easy to sample from, to the posterior conditioned on a truncated observation window. The OT map is obtained as the solution to a stochastic optimization problem that is solved offline using recorded trajectory data from the state and observations. An error analysis of the algorithm is presented under the stationarity and filter stability assumptions, which decomposes the error into two parts related to the truncation window during training and the error due to the optimization procedure. The performance of the proposed method, referred to as optimal transport data-driven filter (OT-DDF), is evaluated for several numerical examples, highlighting its significant computational efficiency during the online stage while maintaining the flexibility and accuracy of OT methods in nonlinear filtering.
comment: 7 pages, 3 figures
Optimal Control of Reduced Left-Invariant Hybrid Control Systems
Optimal control is ubiquitous in many fields of engineering. A common technique to find candidate solutions is via Pontryagin's maximum principle. An unfortunate aspect of this method is that the dimension of system doubles. When the system evolves on a Lie group and the system is invariant under left (or right) translations, Lie-Poisson reduction can be applied to eliminate half of the dimensions (and returning the dimension of the problem to the back to the original number). Hybrid control systems are an extension of (continuous) control systems by allowing for sudden changes to the state. Examples of such systems include the bouncing ball - the velocity instantaneously jumps during a bounce, the thermostat - controls switch to on or off, and a sailboat undergoing tacking. The goal of this work is to extend the idea of Lie-Poisson reduction to the optimal control of these systems. If $n$ is the dimension of the original system, $2n$ is the dimension of the system produced by the maximum principle. In the case of classical Lie-Poisson reduction, the dimension drops back down to $n$. This, unfortunately, is impossible in hybrid systems as there must be an auxiliary variable encoding whether or not an event occurs. As such, the analogous hybrid Lie-Poisson reduction results in a $n+1$ dimensional system. The purpose of this work is to develop and present this technique.
comment: Submitted for a conference. 6 pages
☆ A Triangular Normal Form for x-Flat Control-Affine Two-Input Systems
This paper is devoted to normal forms for x-flat control-affine systems with two inputs. We propose a general triangular normal form which contains several other normal forms discussed in the literature as special cases. We derive conditions under which a system with given x-flat output can be transformed into the proposed triangular form. Based on the triangular form we motivate a simple algorithm for identifying candidates for flat outputs.
☆ Sparse additive function decompositions facing basis transforms
High-dimensional real-world systems can often be well characterized by a small number of simultaneous low-complexity interactions. The analysis of variance (ANOVA) decomposition and the anchored decomposition are typical techniques to find sparse additive decompositions of functions. In this paper, we are interested in a setting, where these decompositions are not directly spare, but become so after an appropriate basis transform. Noting that the sparsity of those additive function decompositions is equivalent to the fact that most of its mixed partial derivatives vanish, we can exploit a connection to the underlying function graphs to determine an orthogonal transform that realizes the appropriate basis change. This is done in three steps: we apply singular value decomposition to minimize the number of vertices of the function graph, and joint block diagonalization techniques of families of matrices followed by sparse minimization based on relaxations of the zero ''norm'' for minimizing the number of edges. For the latter one, we propose and analyze minimization techniques over the manifold of special orthogonal matrices. Various numerical examples illustrate the reliability of our approach for functions having, after a basis transform, a sparse additive decomposition into summands with at most two variables.
comment: 46 pages, 10 figures, 8 tables
♻ ☆ From Complexity to Clarity: Analytical Expressions of Deep Neural Network Weights via Clifford's Geometric Algebra and Convexity
In this paper, we introduce a novel analysis of neural networks based on geometric (Clifford) algebra and convex optimization. We show that optimal weights of deep ReLU neural networks are given by the wedge product of training samples when trained with standard regularized loss. Furthermore, the training problem reduces to convex optimization over wedge product features, which encode the geometric structure of the training dataset. This structure is given in terms of signed volumes of triangles and parallelotopes generated by data vectors. The convex problem finds a small subset of samples via $\ell_1$ regularization to discover only relevant wedge product features. Our analysis provides a novel perspective on the inner workings of deep neural networks and sheds light on the role of the hidden layers.
♻ ☆ Graph-structured tensor optimization for nonlinear density control and mean field games
In this work we develop a numerical method for solving a type of convex graph-structured tensor optimization problems. This type of problems, which can be seen as a generalization of multi-marginal optimal transport problems with graph-structured costs, appear in many applications. Examples are unbalanced optimal transport and multi-species potential mean field games, where the latter is a class of nonlinear density control problems. The method we develop is based on coordinate ascent in a Lagrangian dual, and under mild assumptions we prove that the algorithm converges globally. Moreover, under a set of stricter assumptions, the algorithm converges R-linearly. To perform the coordinate ascent steps one has to compute projections of the tensor, and doing so by brute force is in general not computationally feasible. Nevertheless, for certain graph structures it is possible to derive efficient methods for computing these projections, and here we specifically consider the graph structure that occurs in multi-species potential mean field games. We also illustrate the methodology on a numerical example from this problem class.
comment: 27 pages. Revision: among other things, the section on Convex dynamic network flow problems has been removed, and the section on Multi-species potential mean field games has been expanded
♻ ☆ Reducing non-negativity over general semialgebraic sets to non-negativity over simple sets
A non-negativity certificate (NNC) is a way to write a polynomial so that its non-negativity on a semialgebraic set becomes evident. Positivstellens\"atze (Ps\"atze) guarantee the existence of NNCs. Both, NNCs and Ps\"atze underlie powerful algorithmic techniques for optimization. This paper proposes a universal approach to derive new Ps\"atze for general semialgebraic sets from ones developed for simpler sets, such as a box, a simplex, or the non-negative orthant. We provide several results illustrating the approach. First, by considering Handelman's Positivstellensatz (Psatz) over a box, we construct non-SOS Schm\"{u}dgen-type Ps\"atze over any compact semialgebraic set. That is, a family of Ps\"atze that follow the structure of the fundamental Schm\"{u}dgen's Psatz, but where instead of SOS polynomials, any class of polynomials containing the non-negative constants can be used, such as SONC, DSOS/SDSOS, hyperbolic or sums of AM/GM polynomials. Secondly, by considering the simplex as the simple set, we derive a sparse Psatz over general compact sets, which does not require any structural assumptions of the set. Finally, by considering P\'olya's Psatz over the non-negative orthant, we derive a new non-SOS Psatz over unbounded sets which satisfy some generic conditions. All these results contribute to the literature regarding the use of non-SOS polynomials and sparse NNCs to derive Ps\"atze over compact and unbounded sets. Throughout the article, we illustrate our results with relevant examples and numerical experiments.
comment: 33 pages, 4 tables, 2 figures
♻ ☆ Inferring solar differential rotation and viscosity via passive imaging with inertial waves
The recent discovery of inertial waves on the surface of the Sun offers new possibilities to learn about the solar interior. These waves are long-lived with a period on the order of the Sun rotation period ($\sim$27 days) and are sensitive to parameters deep inside the Sun. They are excited by turbulent convection, leading to a passive imaging problem. In this work, we present the forward and inverse problem of reconstructing viscosity and differential rotation on the Sun from cross-covariance observations of these inertial waves.
comment: proceedings paper
♻ ☆ The role of individual compensation and acceptance decisions in crowdsourced delivery
One of the recent innovations in urban distribution is crowdsourced delivery, where deliveries are made by occasional drivers who wish to utilize their surplus resources (unused transport capacity) by making deliveries in exchange for some compensation. The potential benefits of crowdsourced delivery include reduced delivery costs and increased flexibility (by scaling delivery capacity up and down as needed). The use of occasional drivers poses new challenges because (unlike traditional couriers) neither their availability nor their behavior in accepting delivery offers is certain. The relationship between the compensation offered to occasional drivers and the probability that they will accept a task has been largely neglected in the scientific literature. Therefore, we consider a setting in which compensation-dependent acceptance probabilities are explicitly considered in the process of assigning delivery tasks to occasional drivers. We propose a mixed-integer nonlinear model that minimizes the expected delivery costs while identifying optimal assignments of tasks to a mix of professional and occasional drivers and their compensation. We propose an exact two-stage solution algorithm that allows to decompose compensation and assignment decisions for generic acceptance probability functions and show that the runtime of this algorithm is polynomial under mild conditions. Finally, we also study a more general case of the considered problem setting, show that it is NP-hard and propose an approximate linearization scheme of our mixed-integer nonlinear model. The results of our computational study show clear advantages of our new approach over existing ones. They also indicate that these advantages remain in dynamic settings when tasks and drivers are revealed over time and in which case our method constitutes a fast, yet powerful heuristic.
♻ ☆ Fast Nonlinear Two-Time-Scale Stochastic Approximation: Achieving $O(1/k)$ Finite-Sample Complexity
This paper proposes to develop a new variant of the two-time-scale stochastic approximation to find the roots of two coupled nonlinear operators, assuming only noisy samples of these operators can be observed. Our key idea is to leverage the classic Ruppert-Polyak averaging technique to dynamically estimate the operators through their samples. The estimated values of these averaging steps will then be used in the two-time-scale stochastic approximation updates to find the desired solution. Our main theoretical result is to show that under the strongly monotone condition of the underlying nonlinear operators the mean-squared errors of the iterates generated by the proposed method converge to zero at an optimal rate $O(1/k)$, where $k$ is the number of iterations. Our result significantly improves the existing result of two-time-scale stochastic approximation, where the best known finite-time convergence rate is $O(1/k^{2/3})$. We illustrate this result by applying the proposed method to develop new reinforcement learning algorithms with improved performance.
♻ ☆ Solution-Set Geometry and Regularization Path of a Nonconvexly Regularized Convex Sparse Model
The generalized minimax concave (GMC) penalty is a nonconvex sparse regularizer which can preserve the overall-convexity of the regularized least-squares problem. In this paper, we focus on a significant instance of the GMC model termed scaled GMC (sGMC), and present various notable findings on its solution-set geometry and regularization path. Our investigation indicates that while the sGMC penalty is a nonconvex extension of the LASSO penalty (i.e., the $\ell_1$-norm), the sGMC model preserves many celebrated properties of the LASSO model, hence can serve as a less biased surrogate of LASSO without losing its advantages. Specifically, for a fixed regularization parameter $\lambda$, we show that the solution-set geometry, solution uniqueness and sparseness of the sGMC model can be characterized in a similar elegant way to the LASSO model (see, e.g., Osborne et al. 2000, R. J. Tibshirani 2013). For a varying $\lambda$, we prove that the sGMC solution set is a continuous polytope-valued mapping of $\lambda$. Most noticeably, our study indicates that similar to LASSO, the minimum $\ell_2$-norm regularization path of the sGMC model is continuous and piecewise linear in $\lambda$. Based on these theoretical results, an efficient regularization path algorithm is proposed for the sGMC model, extending the well-known least angle regression (LARS) algorithm for LASSO. We prove the correctness and finite termination of the proposed algorithm under a mild assumption, and confirm its correctness-in-general-situation, efficiency, and practical utility through numerical experiments. Many results in this study also contribute to the theoretical research of LASSO.
comment: 53 pages, 10 figures. Submitted to journal
♻ ☆ Turnpike and dissipativity in generalized discrete-time stochastic linear-quadratic optimal control
We investigate different turnpike phenomena of generalized discrete-time stochastic linear-quadratic optimal control problems. Our analysis is based on a novel strict dissipativity notion for such problems, in which a stationary stochastic process replaces the optimal steady state of the deterministic setting. We show that from this time-varying dissipativity notion, we can conclude turnpike behaviors concerning different objects like distributions, moments, or sample paths of the stochastic system and that the distributions of the stationary pair can be characterized by a stationary optimization problem. The analytical findings are illustrated by numerical simulations.
♻ ☆ Positive Competitive Networks for Sparse Reconstruction
We propose and analyze a continuous-time firing-rate neural network, the positive firing-rate competitive network (\pfcn), to tackle sparse reconstruction problems with non-negativity constraints. These problems, which involve approximating a given input stimulus from a dictionary using a set of sparse (active) neurons, play a key role in a wide range of domains, including for example neuroscience, signal processing, and machine learning. First, by leveraging the theory of proximal operators, we relate the equilibria of a family of continuous-time firing-rate neural networks to the optimal solutions of sparse reconstruction problems. Then, we prove that the \pfcn is a positive system and give rigorous conditions for the convergence to the equilibrium. Specifically, we show that the convergence: (i) only depends on a property of the dictionary; (ii) is linear-exponential, in the sense that initially the convergence rate is at worst linear and then, after a transient, it becomes exponential. We also prove a number of technical results to assess the contractivity properties of the neural dynamics of interest. Our analysis leverages contraction theory to characterize the behavior of a family of firing-rate competitive networks for sparse reconstruction with and without non-negativity constraints. Finally, we validate the effectiveness of our approach via a numerical example.
comment: 26 pages, 9 Figure, 1 Table
♻ ☆ Modeling and Contractivity of Neural-Synaptic Networks with Hebbian Learning
This paper is concerned with the modeling and analysis of two of the most commonly used recurrent neural network models (i.e., Hopfield neural network and firing-rate neural network) with dynamic recurrent connections undergoing Hebbian learning rules. To capture the synaptic sparsity of neural circuits we propose a low dimensional formulation. We then characterize certain key dynamical properties. First, we give biologically-inspired forward invariance results. Then, we give sufficient conditions for the non-Euclidean contractivity of the models. Our contraction analysis leads to stability and robustness of time-varying trajectories -- for networks with both excitatory and inhibitory synapses governed by both Hebbian and anti-Hebbian rules. For each model, we propose a contractivity test based upon biologically meaningful quantities, e.g., neural and synaptic decay rate, maximum in-degree, and the maximum synaptic strength. Then, we show that the models satisfy Dale's Principle. Finally, we illustrate the effectiveness of our results via a numerical example.
comment: 24 pages, 4 figures
♻ ☆ Quantum Langevin Dynamics for Optimization
We initiate the study of utilizing Quantum Langevin Dynamics (QLD) to solve optimization problems, particularly those non-convex objective functions that present substantial obstacles for traditional gradient descent algorithms. Specifically, we examine the dynamics of a system coupled with an infinite heat bath. This interaction induces both random quantum noise and a deterministic damping effect to the system, which nudge the system towards a steady state that hovers near the global minimum of objective functions. We theoretically prove the convergence of QLD in convex landscapes, demonstrating that the average energy of the system can approach zero in the low temperature limit with an exponential decay rate correlated with the evolution time. Numerically, we first show the energy dissipation capability of QLD by retracing its origins to spontaneous emission. Furthermore, we conduct detailed discussion of the impact of each parameter. Finally, based on the observations when comparing QLD with classical Fokker-Plank-Smoluchowski equation, we propose a time-dependent QLD by making temperature and $\hbar$ time-dependent parameters, which can be theoretically proven to converge better than the time-independent case and also outperforms a series of state-of-the-art quantum and classical optimization algorithms in many non-convex landscapes.
comment: 52 pages, 1 table, 25 figures
♻ ☆ Computationally Efficient Chance Constrained Covariance Control with Output Feedback
This paper studies the problem of developing computationally efficient solutions for steering the distribution of the state of a stochastic, linear dynamical system between two boundary Gaussian distributions in the presence of chance-constraints on the state and control input. It is assumed that the state is only partially available through a measurement model corrupted with noise. The filtered state is reconstructed with a Kalman filter, the chance constraints are reformulated as difference of convex (DC) constraints, and the resulting covariance control problem is reformulated as a DC program, which is solved using successive convexification. The efficiency of the proposed method is illustrated on a double integrator example with varying time horizons, and is compared to other state-of-the-art chance constrained covariance control methods.
comment: v2, submitted to CDC '24
♻ ☆ Coherent Equalization of Linear Quantum Systems
This paper introduces a $H_\infty$-like methodology of coherent filtering for equalization of passive linear quantum systems to help mitigate degrading effects of quantum communication channels. For such systems, which include a wide range of linear quantum optical devices and signals, we seek to find a near optimal equalizing filter which is itself a passive quantum system. The problem amounts to solving an optimization problem subject to constraints dictated by the requirement for the equalizer to be physically realizable. By formulating these constraints in the frequency domain, we show that the problem admits a convex $H_\infty$-like formulation. This allows us to derive a set of suboptimal coherent equalizers using $J$-spectral factorization. An additional semidefinite relaxation combined with the Nevanlinna-Pick interpolation is shown to lead to a tractable algorithm for the design of a near optimal coherent equalizer.
♻ ☆ A Traffic Management Framework for On-Demand Urban Air Mobility Systems
Urban Air Mobility (UAM) offers a solution to current traffic congestion by providing on-demand air mobility in urban areas. Effective traffic management is crucial for efficient operation of UAM systems, especially for high-demand scenarios. In this paper, we present a centralized traffic management framework for on-demand UAM systems. Specifically, we provide a scheduling policy, called VertiSync, which schedules the aircraft for either servicing trip requests or rebalancing in the system subject to aircraft safety margins and energy requirements. We characterize the system-level throughput of VertiSync, which determines the demand threshold at which passenger waiting times transition from being stabilized to being increasing over time. We show that the proposed policy is able to maximize throughput for sufficiently large fleet sizes. We demonstrate the performance of VertiSync through a case study for the city of Los Angeles, and show that it significantly reduces passenger waiting times compared to a first-come first-serve scheduling policy.
comment: 9 pages, 6 figures
Systems and Control 53
Control Designs for Critical-Continegency Responsible Grid-Following Inverters and Seamless Transitions To and From Grid-Forming Modes
This article introduces two control frameworks: one for Grid-Following (GFL) inverters aiding Grid-Forming (GFM) inverters in voltage regulation during large contingency events and optimizing power transactions under normal conditions; and another for seamless transitions between grid-tied and grid-isolated setups, managing voltage transient characteristics. In microgrids, GFM inverters regulate voltage, while GFL inverters handle power transactions. The proposed GFL control detects abrupt load/generation changes, adjusting power transactions using local storage to support GFM inverters during contingencies. Additionally, a transition control ensures smooth GFL-GFM shifts, reducing power and voltage fluctuations. Simulation results validate improved voltage regulation during contingencies and enhanced power tracking during slow changes, alongside minimized transient overshoot.
comment: 6 pages, 6 figures, submitted and accepted for 2024 American Control Conference (ACC)
☆ Cascading Blackout Severity Prediction with Statistically-Augmented Graph Neural Networks
Higher variability in grid conditions, resulting from growing renewable penetration and increased incidence of extreme weather events, has increased the difficulty of screening for scenarios that may lead to catastrophic cascading failures. Traditional power-flow-based tools for assessing cascading blackout risk are too slow to properly explore the space of possible failures and load/generation patterns. We add to the growing literature of faster graph-neural-network (GNN)-based techniques, developing two novel techniques for the estimation of blackout magnitude from initial grid conditions. First we propose several methods for employing an initial classification step to filter out safe "non blackout" scenarios prior to magnitude estimation. Second, using insights from the statistical properties of cascading blackouts, we propose a method for facilitating non-local message passing in our GNN models. We validate these two approaches on a large simulated dataset, and show the potential of both to increase blackout size estimation performance.
comment: Accepted to Power Systems Computation Conference (PSCC) 2024
☆ SiMBA: Simplified Mamba-Based Architecture for Vision and Multivariate Time series
Transformers have widely adopted attention networks for sequence mixing and MLPs for channel mixing, playing a pivotal role in achieving breakthroughs across domains. However, recent literature highlights issues with attention networks, including low inductive bias and quadratic complexity concerning input sequence length. State Space Models (SSMs) like S4 and others (Hippo, Global Convolutions, liquid S4, LRU, Mega, and Mamba), have emerged to address the above issues to help handle longer sequence lengths. Mamba, while being the state-of-the-art SSM, has a stability issue when scaled to large networks for computer vision datasets. We propose SiMBA, a new architecture that introduces Einstein FFT (EinFFT) for channel modeling by specific eigenvalue computations and uses the Mamba block for sequence modeling. Extensive performance studies across image and time-series benchmarks demonstrate that SiMBA outperforms existing SSMs, bridging the performance gap with state-of-the-art transformers. Notably, SiMBA establishes itself as the new state-of-the-art SSM on ImageNet and transfer learning benchmarks such as Stanford Car and Flower as well as task learning benchmarks as well as seven time series benchmark datasets. The project page is available on this website ~\url{https://github.com/badripatro/Simba}.
Optimal Exploration Strategy for Regret Minimization in Unconstrained Scalar Optimization Problems IEEE Conference on Decision and Control
We study the problem of determining the optimal exploration strategy in an unconstrained scalar optimization problem depending on an unknown parameter to be learned from online collected noisy data. An optimal trade-off between exploration and exploitation is crucial for effective optimization under uncertainties, and to achieve this we consider a cumulative regret minimization approach over a finite horizon, with each time instant in the horizon characterized by a stochastic exploration signal, whose variance has to be designed. In this setting, under an idealized assumption on an appropriately defined information function associated with the excitation, we are able to show that the optimal exploration strategy is either to use no exploration at all (called lazy exploration) or adding an exploration excitation only at the first time instant of the horizon (called immediate exploration). A quadratic numerical example is used to illustrate the results.
comment: Preprint submitted to IEEE Conference on Decision and Control (CDC 2024)
Optimal Data-Driven Prediction and Predictive Control using Signal Matrix Models
Data-driven control uses a past signal trajectory to characterise the input-output behaviour of a system. Willems' lemma provides a data-based prediction model allowing a control designer to bypass the step of identifying a state-space or transfer function model. This paper provides a more parsimonious formulation of Willems' lemma that separates the model into initial condition matching and predictive control design parts. This avoids the need for regularisers in the predictive control problem that are found in other data-driven predictive control methods. It also gives a closed form expression for the optimal (minimum variance) unbiased predictor of the future output trajectory and applies it for predictive control. Simulation comparisons illustrate very good control performance.
comment: 7 pages, 3 figures. Submitted to IEEE Control Systems Society Letters and the 2024 Conference on Decision and Control
☆ Event-Triggered State Estimation Through Confidence Level
This paper considers the state estimation problem for discrete-time linear systems under event-triggered scheme. In order to improve performance, a novel event-triggered scheme based on confidence level is proposed using the chi-square distribution and mild regularity assumption. In terms of the novel event-triggered scheme, a minimum mean squared error (MMSE) state estimator is proposed using some results presented in this paper. Two algorithms for communication rate estimation of the proposed MMSE state estimator are developed where the first algorithm is based on information with one-step delay, and the second algorithm is based on information with two-step delay. The performance and effectiveness of the proposed MMSE state estimator and the two communication rate estimation algorithms are illustrated using a target tracking scenario.
Control contraction metrics on Lie groups
In this paper, we extend the control contraction metrics (CCM) approach, which was originally proposed for the universal tracking control of nonlinear systems, to those that evolves on Lie groups. Our idea is to view the manifold as a constrained set that is embedded in Euclidean space, and then propose the sufficient conditions for the existence of a CCM and the associated controller design. Notably, we demonstrate that the search for CCM on Lie groups can be reformulated as convex conditions. The results extend the applicability of the CCM approach and provide a framework for analyzing the behavior of control systems with Lie group structures.
☆ On moment relaxations for linear state feedback controller synthesis with non-convex quadratic costs and constraints IEEE Conference on Decision and Control
We present a simple and effective way to account for non-convex costs and constraints~in~state feedback synthesis, and an interpretation for the variables in which state feedback synthesis is typically convex. We achieve this by deriving the controller design using moment matrices of state and input. It turns out that this approach allows the consideration of non-convex constraints by relaxing them as expectation constraints, and that the variables in which state feedback synthesis is typically convexified can be identified with blocks of these moment matrices.
comment: Preprent to be submitted to IEEE Conference on Decision and Control
☆ Robust optimization for adversarial learning with finite sample complexity guarantees
Decision making and learning in the presence of uncertainty has attracted significant attention in view of the increasing need to achieve robust and reliable operations. In the case where uncertainty stems from the presence of adversarial attacks this need is becoming more prominent. In this paper we focus on linear and nonlinear classification problems and propose a novel adversarial training method for robust classifiers, inspired by Support Vector Machine (SVM) margins. We view robustness under a data driven lens, and derive finite sample complexity bounds for both linear and non-linear classifiers in binary and multi-class scenarios. Notably, our bounds match natural classifiers' complexity. Our algorithm minimizes a worst-case surrogate loss using Linear Programming (LP) and Second Order Cone Programming (SOCP) for linear and non-linear models. Numerical experiments on the benchmark MNIST and CIFAR10 datasets show our approach's comparable performance to state-of-the-art methods, without needing adversarial examples during training. Our work offers a comprehensive framework for enhancing binary linear and non-linear classifier robustness, embedding robustness in learning under the presence of adversaries.
☆ Forecasting the load of Parcel Pickup Points using a Markov Jump Process
The growth of e-commerce has resulted in a surge in parcel deliveries, increasing transportation costs and pollution issues. Alternatives to home delivery have emerged, such as the delivery to so-called parcel pick-up points (PUPs), which eliminates delivery failure due to customers not being at home. Nevertheless, parcels reaching overloaded PUPs may need to be redirected to alternative PUPs, sometimes far from the chosen ones, which may generate customer dissatisfaction. Consequently, predicting the PUP load is critical for a PUP management company to infer the availability of PUPs for future orders and better balance parcel flows between PUPs. This paper proposes a new approach to forecasting the PUP load evolution using a Markov jump process that models the parcel life cycle. The latest known status of each parcel is considered to estimate its contribution to the future load of its target PUP. This approach can account for the variability of activity, the various parcel preparation delays by sellers, and the diversity of parcel carriers that may result in different delivery delays. Here, results are provided for predicting the load associated with parcels ordered from online retailers by customers (Business-to-Customer, B2C). The proposed approach is generic and can also be applied to other parcel flows to PUPs, such as second-hand products (Customer-to-Customer, C2C) sent via a PUP network.
☆ Pursuit-Evasion on a Sphere and When It Can Be Considered Flat
In classical works on a planar differential pursuit-evasion game with a faster pursuer, the intercept point resulting from the equilibrium strategies lies on the Apollonius circle. This property was exploited for the construction of the equilibrium strategies for two faster pursuers against one evader. Extensions for planar multiple-pursuer single-evader scenarios have been considered. We study a pursuit-evasion game on a sphere and the relation of the equilibrium intercept point to the Apollonius domain on the sphere. The domain is a generalization of the planar Apollonius circle set. We find a condition resulting in the intercept point belonging to the Apollonius domain, which is the characteristic of the planar game solution. Finally, we use this characteristic to discuss pursuit and evasion strategies in the context of two pursuers and a single slower evader on the sphere and illustrate it using numerical simulations.
comment: 8 Pages, 5 figures, To be submitted to 2024 Conference on Decision and Control in Milan, Italy
☆ Infrastructure-Assisted Collaborative Perception in Automated Valet Parking: A Safety Perspective
Environmental perception in Automated Valet Parking (AVP) has been a challenging task due to severe occlusions in parking garages. Although Collaborative Perception (CP) can be applied to broaden the field of view of connected vehicles, the limited bandwidth of vehicular communications restricts its application. In this work, we propose a BEV feature-based CP network architecture for infrastructure-assisted AVP systems. The model takes the roadside camera and LiDAR as optional inputs and adaptively fuses them with onboard sensors in a unified BEV representation. Autoencoder and downsampling are applied for channel-wise and spatial-wise dimension reduction, while sparsification and quantization further compress the feature map with little loss in data precision. Combining these techniques, the size of a BEV feature map is effectively compressed to fit in the feasible data rate of the NR-V2X network. With the synthetic AVP dataset, we observe that CP can effectively increase perception performance, especially for pedestrians. Moreover, the advantage of infrastructure-assisted CP is demonstrated in two typical safety-critical scenarios in the AVP setting, increasing the maximum safe cruising speed by up to 3m/s in both scenarios.
comment: 7 pages, 7 figures, 4 tables, accepted by IEEE VTC2024-Spring
☆ Hybrid integrator-gain system based integral resonant controllers for negative imaginary systems
We introduce a hybrid control system called a hybrid integrator-gain system (HIGS) based integral resonant controller (IRC) to stabilize negative imaginary (NI) systems. A HIGS-based IRC has a similar structure to an IRC, with the integrator replaced by a HIGS. We show that a HIGS-based IRC is an NI system. Also, for a SISO NI system with a minimal realization, we show there exists a HIGS-based IRC such that their closed-loop interconnection is asymptotically stable. Also, we propose a proportional-integral-double-integral resonant controller and a HIGS-based proportional-integral-double-integral resonant controller and show that both of them can be applied to asymptotically stabilize an NI system. An example is provided to illustrate the proposed results.
comment: 9 pages, 9 figures
☆ Collision Avoidance Safety Filter for an Autonomous E-Scooter using Ultrasonic Sensors
In this paper, we propose a collision avoidance safety filter for autonomous electric scooters to enable safe operation of such vehicles in pedestrian areas. In particular, we employ multiple low-cost ultrasonic sensors to detect a wide range of possible obstacles in front of the e-scooter. Based on possibly faulty distance measurements, we design a filter to mitigate measurement noise and missing values as well as a gain-scheduled controller to limit the velocity commanded to the e-scooter when required due to imminent collisions. The proposed controller structure is able to prevent collisions with unknown obstacles by deploying a reduced safe velocity ensuring a sufficiently large safety distance. The collision avoidance approach is designed such that it may be easily deployed in similar applications of general micromobility vehicles. The effectiveness of our proposed safety filter is demonstrated in real-world experiments.
☆ Set-membership target search and tracking within an unknown cluttered area using cooperating UAVs equipped with vision systems
This paper addresses the problem of target search and tracking using a fleet of cooperating UAVs evolving in some unknown region of interest containing an a priori unknown number of moving ground targets. Each drone is equipped with an embedded Computer Vision System (CVS), providing an image with labeled pixels and a depth map of the observed part of its environment. Moreover, a box containing the corresponding pixels in the image frame is available when a UAV identifies a target. Hypotheses regarding information provided by the pixel classification, depth map construction, and target identification algorithms are proposed to allow its exploitation by set-membership approaches. A set-membership target location estimator is developed using the information provided by the CVS. Each UAV evaluates sets guaranteed to contain the location of the identified targets and a set possibly containing the locations of targets still to be identified. Then, each UAV uses these sets to search and track targets cooperatively.
☆ Improved Long Short-Term Memory-based Wastewater Treatment Simulators for Deep Reinforcement Learning
Even though Deep Reinforcement Learning (DRL) showed outstanding results in the fields of Robotics and Games, it is still challenging to implement it in the optimization of industrial processes like wastewater treatment. One of the challenges is the lack of a simulation environment that will represent the actual plant as accurately as possible to train DRL policies. Stochasticity and non-linearity of wastewater treatment data lead to unstable and incorrect predictions of models over long time horizons. One possible reason for the models' incorrect simulation behavior can be related to the issue of compounding error, which is the accumulation of errors throughout the simulation. The compounding error occurs because the model utilizes its predictions as inputs at each time step. The error between the actual data and the prediction accumulates as the simulation continues. We implemented two methods to improve the trained models for wastewater treatment data, which resulted in more accurate simulators: 1- Using the model's prediction data as input in the training step as a tool of correction, and 2- Change in the loss function to consider the long-term predicted shape (dynamics). The experimental results showed that implementing these methods can improve the behavior of simulators in terms of Dynamic Time Warping throughout a year up to 98% compared to the base model. These improvements demonstrate significant promise in creating simulators for biological processes that do not need pre-existing knowledge of the process but instead depend exclusively on time series data obtained from the system.
☆ Direct and Indirect Hydrogen Storage: Dynamics and Interactions in the Transition to a Renewable Energy Based System for Europe
To move towards a low-carbon society by 2050, understanding the intricate dynamics of energy systems is critical. Our study examines these interactions through the lens of hydrogen storage, dividing it into 'direct' and 'indirect' hydrogen storage. Direct hydrogen storage involves electrolysis-produced hydrogen being stored before use, while indirect storage first transforms hydrogen into gas via the Sabatier process for later energy distribution. Firstly, we utilize the PyPSA-Eur-Sec-30-path model to capture the interactions within the energy system. The model is an hour-level, one node per country system that encompasses a range of energy transformation technologies, outlining a pathway for Europe to reduce carbon emissions by 95 percent by 2050 compared to 1990, with updates every 5 years. Subsequently, we employ both quantitative and qualitative approaches to thoroughly analyze these complex relationships. Our research indicates that during the European green transition, cross-country flow of electricity will play an important role in Europe's rapid decarbonization stage before the large-scale introduction of energy storage. Under the paper cost assumptions, fuel cells are not considered a viable option. This research further identifies the significant impact of natural resource variability on the local energy mix, highlighting indirect hydrogen storage as a common solution due to the better economic performance and actively fluctuation pattern. Specifically, indirect hydrogen storage will contribute at least 60 percent of hydrogen storage benefits, reaching 100 percent in Italy. Moreover, its fluctuation pattern will change with the local energy structure, which is a distinct difference with the unchanged pattern of direct hydrogen storage and battery storage.
☆ A Twin Delayed Deep Deterministic Policy Gradient Algorithm for Autonomous Ground Vehicle Navigation via Digital Twin Perception Awareness
Autonomous ground vehicle (UGV) navigation has the potential to revolutionize the transportation system by increasing accessibility to disabled people, ensure safety and convenience of use. However, UGV requires extensive and efficient testing and evaluation to ensure its acceptance for public use. This testing are mostly done in a simulator which result to sim2real transfer gap. In this paper, we propose a digital twin perception awareness approach for the control of robot navigation without prior creation of the virtual environment (VT) environment state. To achieve this, we develop a twin delayed deep deterministic policy gradient (TD3) algorithm that ensures collision avoidance and goal-based path planning. We demonstrate the performance of our approach on different environment dynamics. We show that our approach is capable of efficiently avoiding collision with obstacles and navigating to its desired destination, while at the same time safely avoids obstacles using the information received from the LIDAR sensor mounted on the robot. Our approach bridges the gap between sim-to-real transfer and contributes to the adoption of UGVs in real world. We validate our approach in simulation and a real-world application in an office space.
comment: 8 pages, 7 figures
☆ Implementation of Firm-Dispatchable Generation in South Africa
South Africa is currently facing a critical situation in its power generation landscape, which is plagued by frequent power outages and the need to move from fossil fuels to renewable energy sources. This period emphasizes the importance of having firm-dispatchable power to balance out the intermittent nature of wind and solar energy sources. The paper proposes to repurpose old coal-fired power plants to generate firm-dispatchable energy in line with the principles of a Just Transition. Eskom's coal plants are approaching the end of their economic life, and their declining energy availability factor is becoming a challenge in meeting the country's energy needs. The study suggests that a comprehensive strategy that integrates wind, solar, and firm-dispatchable power can be cost-effective and reliable compared to the traditional coal-based approach or the nuclear alternative. The study emphasizes the necessity of a 25-year plan that would invest in flexible and modular dispatchable generation. It also highlights the strategic location of this generating capacity, including repurposing decommissioned coal plant sites. The proposed model integrates private investment, adheres to established best practices, and emphasizes adaptability to changing demand dynamics. The study provides a roadmap for enabling firm-dispatchable capacity for South Africa's energy transition, emphasizing economic prudence, environmental sustainability, and alignment with the principles of the Just Transition program.
☆ On the Solution Uniqueness of Data-Driven Modeling of Flexible Loads
This letter first explores the solution uniqueness of the data-driven modeling of price-responsive flexible loads (PFL). The PFL on the demand side is critical in modern power systems. An accurate PFL model is fundamental for system operations. Yet, whether the PFL model can be uniquely and correctly identified from operational data remains unclear. To address this, we analyze the structural and practical identifiability of the PFL model, deriving the condition for the solution uniqueness. Based on this, we point out the implications for selecting physical models of PFL to enhance the identification results. Numerical results validate this work.
☆ Linear Quadratic Guidance Law for Joint Motion Planning of a Pursuer-Turret Assembly
This paper presents joint motion planning of a vehicle with an attached rotating turret. The turret has a limited range as well as the field of view. The objective is capture a maneuvering target such that at the terminal time it is withing the field-of-view and range limits. Catering to it, we present a minimum effort guidance law that commensurate for the turn rate abilities of the vehicle and the turret. The guidance law is obtained using linearization about the collision triangle and admits an analytical solution. Simulation results are presented to exemplify the cooperation between the turret and the vehicle.
☆ Real-time Safety Index Adaptation for Parameter-varying Systems via Determinant Gradient Ascend
Safety Index Synthesis (SIS) is critical for deriving safe control laws. Recent works propose to synthesize a safety index (SI) via nonlinear programming and derive a safe control law such that the system 1) achieves forward invariant (FI) with some safe set and 2) guarantees finite time convergence (FTC) to that safe set. However, real-world system dynamics can vary during run-time, making the control law infeasible and invalidating the initial SI. Since the full SIS nonlinear programming is computationally expensive, it is infeasible to re-synthesize the SI each time the dynamics are perturbed. To address that, this paper proposes an efficient approach to adapting the SI to varying system dynamics and maintaining the feasibility of the safe control law. The proposed method leverages determinant gradient ascend and derives a closed-form update to safety index parameters, enabling real-time adaptation performance. A numerical study validates the effectiveness of our approach.
comment: Accepted to American Control Conference (ACC) 2024
☆ Data-Driven Predictive Control with Adaptive Disturbance Attenuation for Constrained Systems
In this paper, we propose a novel data-driven predictive control approach for systems subject to time-domain constraints. The approach combines the strengths of H-infinity control for rejecting disturbances and MPC for handling constraints. In particular, the approach can dynamically adapt H-infinity disturbance attenuation performance depending on measured system state and forecasted disturbance level to satisfy constraints. We establish theoretical properties of the approach including robust guarantees of closed-loop stability, disturbance attenuation, constraint satisfaction under noisy data, as well as sufficient conditions for recursive feasibility, and illustrate the approach with a numerical example.
comment: 11 pages, 2 figures
☆ Structured stability analysis of networked systems with uncertain links
An input-output approach to stability analysis is explored for networked systems with uncertain link dynamics. The main result consists of a collection of integral quadratic constraints, which together imply robust stability of the uncertain networked system, under the assumption that stability is achieved with ideal links. The conditions are decentralized inasmuch as each involves only agent and uncertainty model parameters that are local to a corresponding link. This makes the main result, which imposes no restriction on network structure, suitable for the study of large-scale systems.
☆ Network Learning with Directional Sign Patterns
Complex systems can be effectively modeled via graphs that encode networked interactions, where relations between entities or nodes are often quantified by signed edge weights, e.g., promotion/inhibition in gene regulatory networks, or encoding political of friendship differences in social networks. However, it is often the case that only an aggregate consequence of such edge weights that characterize relations may be directly observable, as in protein expression of in gene regulatory networks. Thus, learning edge weights poses a significant challenge that is further exacerbated for intricate and large-scale networks. In this article, we address a model problem to determine the strength of sign-indefinite relations that explain marginal distributions that constitute our data. To this end, we develop a paradigm akin to that of the Schr\"odinger bridge problem and an efficient Sinkhorn type algorithm (more properly, Schr\"odinger-Fortet-Sinkhorn algorithm) that allows fast convergence to parameters that minimize a relative entropy/likelihood criterion between the sought signed adjacency matrix and a prior. The formalism that we present represents a novel generalization of the earlier Schr\"odinger formalism in that marginal computations may incorporate weights that model directionality in underlying relations, and further, that it can be extended to high-order networks -- the Schr\"odinger-Fortet-Sinkhorn algorithm that we derive is applicable all the same and allows geometric convergence to a sought sign-indefinite adjacency matrix or tensor, for high-order networks. We demonstrate our framework with synthetic and real-world examples.
☆ Optimisation of photodetectors design: comparison between Montecarlo and Genetic Algorithms
We present Montecarlo and Genetic Algorithm optimisations applied to the design of photodetectors based on a transimpedance amplifier and a photodiode. The circuit performance is evaluated with a merit function and the systematic search method is used as a reference. The design parameters are the feedback network components and the photodiode bias voltage. To evaluate the optimisations, we define the relative difference between its merit and the optimum merit obtained by the systematic search. In both algorithms, the relative difference decreases with the number of evaluations, following a power law. The power-law exponent for the Genetic Algorithm is larger than that of Montecarlo (0.74 vs. 0.50). We conclude that both algorithms are advantageous compared to the systematic search method, and that the Genetic Algorithm shows a better performance than Montecarlo.
comment: 10 pages, 14 figures
☆ On the Variational Interpretation of Mirror Play in Monotone Games
Mirror play (MP) is a well-accepted primal-dual multi-agent learning algorithm where all agents simultaneously implement mirror descent in a distributed fashion. The advantage of MP over vanilla gradient play lies in its usage of mirror maps that better exploit the geometry of decision domains. Despite extensive literature dedicated to the asymptotic convergence of MP to equilibrium, the understanding of the finite-time behavior of MP before reaching equilibrium is still rudimentary. To facilitate the study of MP's non-equilibrium performance, this work establishes an equivalence between MP's finite-time primal-dual path (mirror path) in monotone games and the closed-loop Nash equilibrium path of a finite-horizon differential game, referred to as mirror differential game (MDG). Our construction of MDG rests on the Brezis-Ekeland variational principle, and the stage cost functional for MDG is Fenchel coupling between MP's iterates and associated gradient updates. The variational interpretation of mirror path in static games as the equilibrium path in MDG holds in deterministic and stochastic cases. Such a variational interpretation translates the non-equilibrium studies of learning dynamics into a more tractable equilibrium analysis of dynamic games, as demonstrated in a case study on the Cournot game, where MP dynamics corresponds to a linear quadratic game.
☆ Uncertainty Propagation in Stochastic Systems via Mixture Models with Error Quantification
Uncertainty propagation in non-linear dynamical systems has become a key problem in various fields including control theory and machine learning. In this work we focus on discrete-time non-linear stochastic dynamical systems. We present a novel approach to approximate the distribution of the system over a given finite time horizon with a mixture of distributions. The key novelty of our approach is that it not only provides tractable approximations for the distribution of a non-linear stochastic system, but also comes with formal guarantees of correctness. In particular, we consider the total variation (TV) distance to quantify the distance between two distributions and derive an upper bound on the TV between the distribution of the original system and the approximating mixture distribution derived with our framework. We show that in various cases of interest, including in the case of Gaussian noise, the resulting bound can be efficiently computed in closed form. This allows us to quantify the correctness of the approximation and to optimize the parameters of the resulting mixture distribution to minimize such distance. The effectiveness of our approach is illustrated on several benchmarks from the control community.
☆ Transactive Local Energy Markets Enable Community-Level Resource Coordination Using Individual Rewards
ALEX (Autonomous Local Energy eXchange) is an economy-driven, transactive local energy market where each participating building is represented by a rational agent. Relying solely on building-level information, this agent minimizes its electricity bill by automating distributed energy resource utilization and trading. This study examines ALEX's capabilities to align participant and grid-stakeholder interests and assesses ALEX's impact on short- and long-term intermittence using a set of community net-load metrics, such as ramping rate, load factor, and peak load. The policies for ALEX's rational agents are generated using dynamic programming through value iteration in conjunction with iterative best response. This facilitates comparing ALEX and a benchmark energy management system, which optimizes building-level self-consumption, ramping rate, and peak net load. Simulations are performed using the open-source CityLearn2022 dataset to provide a pathway for benchmarking by future studies. The experiments demonstrate that ALEX enables the coordination of distributed energy resources across the community. Remarkably, this community-level coordination occurs even though the system is populated by agents who only access building-level information and selfishly maximize their own relative profit. Compared to the benchmark energy management system, ALEX improves across all metrics.
comment: Preprint, submitted to IEEE Access
☆ Balancing Fairness and Efficiency in Energy Resource Allocations
Bringing fairness to energy resource allocation remains a challenge, due to the complexity of system structures and economic interdependencies among users and system operators' decision-making. The rise of distributed energy resources has introduced more diverse heterogeneous user groups, surpassing the capabilities of traditional efficiency-oriented allocation schemes. Without explicitly bringing fairness to user-system interaction, this disparity often leads to disproportionate payments for certain user groups due to their utility formats or group sizes. Our paper addresses this challenge by formalizing the problem of fair energy resource allocation and introducing the framework for aggregators. This framework enables optimal fairness-efficiency trade-offs by selecting appropriate objectives in a principled way. By jointly optimizing over the total resources to allocate and individual allocations, our approach reveals optimized allocation schemes that lie on the Pareto front, balancing fairness and efficiency in resource allocation strategies.
☆ Adaptive Dual Covariance Steering with Active Parameter Estimation
This work examines the optimal covariance steering problem for systems subject to unknown parameters that enter multiplicatively with the state and control, in addition to additive disturbances. In contrast to existing works, the unknown parameters are modeled as random variables and are estimated online. This work proposes the utilization of recursive least squares estimation for efficient parameter identification. A dual control problem is formulated in which the effect of the planned control policy on the parameter estimates is modeled and optimized for. The parameter estimates are then used to modify the pre-computed control policy online in an adaptive control fashion. Finally, the proposed approach is demonstrated in a vehicle control example with closed-loop parameter identification.
☆ Autonomous Driving With Perception Uncertainties: Deep-Ensemble Based Adaptive Cruise Control
Autonomous driving depends on perception systems to understand the environment and to inform downstream decision-making. While advanced perception systems utilizing black-box Deep Neural Networks (DNNs) demonstrate human-like comprehension, their unpredictable behavior and lack of interpretability may hinder their deployment in safety critical scenarios. In this paper, we develop an Ensemble of DNN regressors (Deep Ensemble) that generates predictions with quantification of prediction uncertainties. In the scenario of Adaptive Cruise Control (ACC), we employ the Deep Ensemble to estimate distance headway to the lead vehicle from RGB images and enable the downstream controller to account for the estimation uncertainty. We develop an adaptive cruise controller that utilizes Stochastic Model Predictive Control (MPC) with chance constraints to provide a probabilistic safety guarantee. We evaluate our ACC algorithm using a high-fidelity traffic simulator and a real-world traffic dataset and demonstrate the ability of the proposed approach to effect speed tracking and car following while maintaining a safe distance headway. The out-of-distribution scenarios are also examined.
♻ ☆ Empowering Autonomous Driving with Large Language Models: A Safety Perspective ICLR2024
Autonomous Driving (AD) encounters significant safety hurdles in long-tail unforeseen driving scenarios, largely stemming from the non-interpretability and poor generalization of the deep neural networks within the AD system, particularly in out-of-distribution and uncertain data. To this end, this paper explores the integration of Large Language Models (LLMs) into AD systems, leveraging their robust common-sense knowledge and reasoning abilities. The proposed methodologies employ LLMs as intelligent decision-makers in behavioral planning, augmented with a safety verifier shield for contextual safety learning, for enhancing driving performance and safety. We present two key studies in a simulated environment: an adaptive LLM-conditioned Model Predictive Control (MPC) and an LLM-enabled interactive behavior planning scheme with a state machine. Demonstrating superior performance and safety metrics compared to state-of-the-art approaches, our approach shows the promising potential for using LLMs for autonomous vehicles.
comment: Accepted to LLMAgent workshop @ICLR2024
♻ ☆ A Benchmark for the Application of Distributed Control Techniques to the Electricity Network of the European Economic Area
The European Economic Area Electricity Network Benchmark (EEA-ENB) is a multi-area power system representing the European network of transmission systems for electricity to facilitate the application of distributed control techniques. In the EEA-ENB we consider the Load Frequency Control (LFC) problem in the presence of renewable energy sources (RESs), and energy storage systems (ESSs). RESs are known to cause instability in power networks due to their inertia-less and intermittent characteristics, while ESSs are introduced as a resource to mitigate the problem. In the EEA-ENB, particular attention is dedicated to Distributed Model Predictive Control (DMPC), whose application is often limited to small and homogeneous test cases due to the lack of standardized large-scale scenarios for testing, and due to the large computation time required to obtain a centralized MPC action for performance comparison with DMPC strategies under consideration. The second problem is exacerbated when the scale of the system grows. To address these challenges and to provide a real-world-based and control-independent benchmark, the EEA-ENB has been developed. The benchmark includes a centralized MPC strategy providing performance and computation time metrics to compare distributed control within a repeatable and realistic simulation environment.
comment: Updated reference list with reference and DOI to software sources to run the benchmark
♻ ☆ Novelty Detection in Reinforcement Learning with World Models
Reinforcement learning (RL) using world models has found significant recent successes. However, when a sudden change to world mechanics or properties occurs then agent performance and reliability can dramatically decline. We refer to the sudden change in visual properties or state transitions as novelties. Implementing novelty detection within generated world model frameworks is a crucial task for protecting the agent when deployed. In this paper, we propose straightforward bounding approaches to incorporate novelty detection into world model RL agents, by utilizing the misalignment of the world model's hallucinated states and the true observed states as an anomaly score. We provide effective approaches to detecting novelties in a distribution of transitions learned by an agent in a world model. Finally, we show the advantage of our work in a novel environment compared to traditional machine learning novelty detection methods as well as currently accepted RL focused novelty detection algorithms.
♻ ☆ Graph-structured tensor optimization for nonlinear density control and mean field games
In this work we develop a numerical method for solving a type of convex graph-structured tensor optimization problems. This type of problems, which can be seen as a generalization of multi-marginal optimal transport problems with graph-structured costs, appear in many applications. Examples are unbalanced optimal transport and multi-species potential mean field games, where the latter is a class of nonlinear density control problems. The method we develop is based on coordinate ascent in a Lagrangian dual, and under mild assumptions we prove that the algorithm converges globally. Moreover, under a set of stricter assumptions, the algorithm converges R-linearly. To perform the coordinate ascent steps one has to compute projections of the tensor, and doing so by brute force is in general not computationally feasible. Nevertheless, for certain graph structures it is possible to derive efficient methods for computing these projections, and here we specifically consider the graph structure that occurs in multi-species potential mean field games. We also illustrate the methodology on a numerical example from this problem class.
comment: 27 pages. Revision: among other things, the section on Convex dynamic network flow problems has been removed, and the section on Multi-species potential mean field games has been expanded
♻ ☆ Inferentialist Resource Semantics
In systems modelling, a system typically comprises located resources relative to which processes execute. One important use of logic in informatics is in modelling such systems for the purpose of reasoning (perhaps automated) about their behaviour and properties. To this end, one requires an interpretation of logical formulae in terms of the resources and states of the system; such an interpretation is called a resource semantics of the logic. This paper shows how inferentialism -- the view that meaning is given in terms of inferential behaviour -- enables a versatile and expressive framework for resource semantics. Specifically, how inferentialism seamlessly incorporates the assertion-based approach of the logic of Bunched Implications, foundational in program verification (e.g., as the basis of Separation Logic), and the renowned number-of-uses reading of Linear Logic. This integration enables reasoning about shared and separated resources in intuitive and familiar ways, as well as about the composition and interfacing of system components.
♻ ☆ Sparse Mean Field Load Balancing in Large Localized Queueing Systems
Scalable load balancing algorithms are of great interest in cloud networks and data centers, necessitating the use of tractable techniques to compute optimal load balancing policies for good performance. However, most existing scalable techniques, especially asymptotically scaling methods based on mean field theory, have not been able to model large queueing networks with strong locality. Meanwhile, general multi-agent reinforcement learning techniques can be hard to scale and usually lack a theoretical foundation. In this work, we address this challenge by leveraging recent advances in sparse mean field theory to learn a near-optimal load balancing policy in sparsely connected queueing networks in a tractable manner, which may be preferable to global approaches in terms of wireless communication overhead. Importantly, we obtain a general load balancing framework for a large class of sparse bounded-degree wireless topologies. By formulating a novel mean field control problem in the context of graphs with bounded degree, we reduce the otherwise difficult multi-agent problem to a single-agent problem. Theoretically, the approach is justified by approximation guarantees. Empirically, the proposed methodology performs well on several realistic and scalable wireless network topologies as compared to a number of well-known load balancing heuristics and existing scalable multi-agent reinforcement learning methods.
♻ ☆ A Control-Recoverable Added-Noise-based Privacy Scheme for LQ Control in Networked Control Systems
As networked control systems continue to evolve, ensuring the privacy of sensitive data becomes an increasingly pressing concern, especially in situations where the controller is physically separated from the plant. In this paper, we propose a secure control scheme for computing linear quadratic control in a networked control system utilizing two networked controllers, a privacy encoder and a control restorer. Specifically, the encoder generates two state signals blurred with random noise and sends them to the controllers, while the restorer reconstructs the correct control signal. The proposed design effectively preserves the privacy of the control system's state without sacrificing the control performance. We theoretically quantify the privacy-preserving performance in terms of the state estimation error of the controllers and the disclosure probability. Additionally, the proposed privacy-preserving scheme is also proven to satisfy differential privacy. Moreover, we extend the proposed privacy-preserving scheme and evaluation method to cases where collusion between two controllers occurs. Finally, we verify the validity of our proposed scheme through simulations.
♻ ☆ End-to-End Reinforcement Learning of Koopman Models for Economic Nonlinear Model Predictive Control
(Economic) nonlinear model predictive control ((e)NMPC) requires dynamic models that are sufficiently accurate and computationally tractable. Data-driven surrogate models for mechanistic models can reduce the computational burden of (e)NMPC; however, such models are typically trained by system identification for maximum prediction accuracy on simulation samples and perform suboptimally in (e)NMPC. We present a method for end-to-end reinforcement learning of Koopman surrogate models for optimal performance as part of (e)NMPC. We apply our method to two applications derived from an established nonlinear continuous stirred-tank reactor model. The controller performance is compared to that of (e)NMPCs utilizing models trained using system identification, and model-free neural network controllers trained using reinforcement learning. We show that the end-to-end trained models outperform those trained using system identification in (e)NMPC, and that, in contrast to the neural network controllers, the (e)NMPC controllers can react to changes in the control setting without retraining.
comment: manuscript (18 pages, 7 figures, 5 tables), supplementary materials (3 pages, 2 tables)
♻ ☆ Positive Competitive Networks for Sparse Reconstruction
We propose and analyze a continuous-time firing-rate neural network, the positive firing-rate competitive network (\pfcn), to tackle sparse reconstruction problems with non-negativity constraints. These problems, which involve approximating a given input stimulus from a dictionary using a set of sparse (active) neurons, play a key role in a wide range of domains, including for example neuroscience, signal processing, and machine learning. First, by leveraging the theory of proximal operators, we relate the equilibria of a family of continuous-time firing-rate neural networks to the optimal solutions of sparse reconstruction problems. Then, we prove that the \pfcn is a positive system and give rigorous conditions for the convergence to the equilibrium. Specifically, we show that the convergence: (i) only depends on a property of the dictionary; (ii) is linear-exponential, in the sense that initially the convergence rate is at worst linear and then, after a transient, it becomes exponential. We also prove a number of technical results to assess the contractivity properties of the neural dynamics of interest. Our analysis leverages contraction theory to characterize the behavior of a family of firing-rate competitive networks for sparse reconstruction with and without non-negativity constraints. Finally, we validate the effectiveness of our approach via a numerical example.
comment: 26 pages, 9 Figure, 1 Table
♻ ☆ Robust Direct Data-Driven Control for Probabilistic Systems
We propose a data-driven control method for systems with aleatoric uncertainty, for example, robot fleets with variations between agents. Our method leverages shared trajectory data to increase the robustness of the designed controller and thus facilitate transfer to new variations without the need for prior parameter and uncertainty estimations. In contrast to existing work on experience transfer for performance, our approach focuses on robustness and uses data collected from multiple realizations to guarantee generalization to unseen ones. Our method is based on scenario optimization combined with recent formulations for direct data-driven control. We derive lower bounds on the amount of data required to achieve quadratic stability for probabilistic systems with aleatoric uncertainty and demonstrate the benefits of our data-driven method through a numerical example. We find that the learned controllers generalize well to high variations in the dynamics even when based on only a few short open-loop trajectories. Robust experience transfer enables the design of safe and robust controllers that work out of the box without any additional learning during deployment.
♻ ☆ Performance Investigation of an Optimal Control Strategy for Zero-Emission Operations of Shipboard Microgrids
This work introduces an efficient power management approach for shipboard microgrids that integrates diesel generators, a fuel cell, and battery energy storage system. This strategy addresses both unit commitment and power dispatch, considering the zero-emission capability of the ship, as well as optimizing the ship's speed. The optimization is done through mixed integer linear programming with the objective of minimizing the operational cost of all the power resources. Evaluations are conducted on a notional all-electric ship, with electrical load simulated using a Markov chain based on actual measurement data. The findings underscore the effectiveness of the proposed strategy in optimizing fuel consumption while ensuring protection against blackout occurrences.
comment: Submitted to SPEEDAM 2024
♻ ☆ A Wind-Aware Path Planning Method for UAV-Asisted Bridge Inspection
In response to the gap in considering wind conditions in the bridge inspection using unmanned aerial vehicle (UAV) , this paper proposes a path planning method for UAVs that takes into account the influence of wind, based on the simulated annealing algorithm. The algorithm considers the wind factors, including the influence of different wind speeds and directions at the same time on the path planning of the UAV. Firstly, An environment model is constructed specifically for UAV bridge inspection, taking into account the various objective functions and constraint conditions of UAVs. A more sophisticated and precise mathematical model is then developed based on this environmental model to enable efficient and effective UAV path planning. Secondly, the bridge separation planning model is applied in a novel way, and a series of parameters are simulated, including the adjustment of the initial temperature value. The experimental results demonstrate that, compared with traditional local search algorithms, the proposed method achieves a cost reduction of 30.05\% and significantly improves effectiveness. Compared to path planning methods that do not consider wind factors, the proposed approach yields more realistic and practical results for UAV applications, as demonstrated by its improved effectiveness in simulations. These findings highlight the value of our method in facilitating more accurate and efficient UAV path planning in wind-prone environments.
comment: After carefully analysis, there is a bit design flaws in Algorithm 1. The experimental work of the paper is not comprehensive,which lacks an evaluation of the algorithm's running time
♻ ☆ DRAMS: Double-RIS Assisted Multihop Routing Scheme for Device-to-Device Communication
Reconfigurable intelligent surfaces (RISs) is a promising solution for enhancing the performance of multihop wireless communication networks. In this paper, we propose a double-RIS assisted multihop routing scheme for a device-to-device (D2D) communication network. Specifically, the scheme is dependent on the already deployed RISs and users in the surroundings. Besides the RISs, the emphasis of this work is to make more use of the existing intermediate users (IUs), which can act as relays. Hence, the density of RIS deployment in the surroundings can be reduced, which leads to the avoidance of resource wastage. However, we cannot solely depend on the IUs because this implies complete dependence on their availability for relaying and as a result, the aspect of reliability in terms of delay-constrained information transfer cannot be guaranteed. Moreover, the IUs are considered capable of energy harvesting and as a result, they do not waste their own energy in the process of volunteering to act as a relay for other users. Numerical results demonstrate the advantage of the proposed scheme over some existing approaches and lastly, useful insights related to the scheme design are also drawn, where we characterize the maximum acceptable delay at each hop under different set-ups.
comment: To appear in Elsevier Computer Communications
♻ ☆ An Optimal Control Framework for Influencing Human Driving Behavior in Mixed-Autonomy Traffic
As autonomous vehicles (AVs) become increasingly prevalent, their interaction with human drivers presents a critical challenge. Current AVs lack social awareness, causing behavior that is often awkward or unsafe. To combat this, social AVs, which are proactive rather than reactive in their behavior, have been explored in recent years. With knowledge of robot-human interaction dynamics, a social AV can influence a human driver to exhibit desired behaviors by strategically altering its own behaviors. In this paper, we present a novel framework for achieving human influence. The foundation of our framework lies in an innovative use of control barrier functions to formulate the desired objectives of influence as constraints in an optimal control problem. The computed controls gradually push the system state toward satisfaction of the objectives, e.g. slowing the human down to some desired speed. We demonstrate the proposed framework's feasibility in a variety of scenarios related to car-following and lane changes, including multi-robot and multi-human configurations. In two case studies, we validate the framework's effectiveness when applied to the problems of traffic flow optimization and aggressive behavior mitigation. Given these results, the main contribution of our framework is its versatility in a wide spectrum of influence objectives and mixed-autonomy configurations.
comment: Accepted to American Control Conference (ACC) 2024
♻ ☆ Learning Hierarchical Control For Multi-Agent Capacity-Constrained Systems
This paper introduces a novel data-driven hierarchical control scheme for managing a fleet of nonlinear, capacity-constrained autonomous agents in an iterative environment. We propose a control framework consisting of a high-level dynamic task assignment and routing layer and low-level motion planning and tracking layer. Each layer of the control hierarchy uses a data-driven Model Predictive Control (MPC) policy, maintaining bounded computational complexity at each calculation of a new task assignment or actuation input. We utilize collected data to iteratively refine estimates of agent capacity usage, and update MPC policy parameters accordingly. Our approach leverages tools from iterative learning control to integrate learning at both levels of the hierarchy, and coordinates learning between levels in order to maintain closed-loop feasibility and performance improvement of the connected architecture.
♻ ☆ A Survey of Predictive Maintenance: Systems, Purposes and Approaches
This paper highlights the importance of maintenance techniques in the coming industrial revolution, reviews the evolution of maintenance techniques, and presents a comprehensive literature review on the latest advancement of maintenance techniques, i.e., Predictive Maintenance (PdM), with emphasis on system architectures, optimization objectives, and optimization methods. In industry, any outages and unplanned downtime of machines or systems would degrade or interrupt a company's core business, potentially resulting in significant penalties and immeasurable reputation and economic loss. Existing traditional maintenance approaches, such as Reactive Maintenance (RM) and Preventive Maintenance (PM), suffer from high prevent and repair costs, inadequate or inaccurate mathematical degradation processes, and manual feature extraction. The incoming fourth industrial revolution is also demanding for a new maintenance paradigm to reduce the maintenance cost and downtime, and increase system availability and reliability. Predictive Maintenance (PdM) is envisioned the solution. In this survey, we first provide a high-level view of the PdM system architectures including PdM 4.0, Open System Architecture for Condition Based Monitoring (OSA-CBM), and cloud-enhanced PdM system. Then, we review the specific optimization objectives, which mainly comprise cost minimization, availability/reliability maximization, and multi-objective optimization. Furthermore, we present the optimization methods to achieve the aforementioned objectives, which include traditional Machine Learning (ML) based and Deep Learning (DL) based approaches. Finally, we highlight the future research directions that are critical to promote the application of DL techniques in the context of PdM.
comment: 38 pages, 23 figures
♻ ☆ Computationally Efficient Chance Constrained Covariance Control with Output Feedback
This paper studies the problem of developing computationally efficient solutions for steering the distribution of the state of a stochastic, linear dynamical system between two boundary Gaussian distributions in the presence of chance-constraints on the state and control input. It is assumed that the state is only partially available through a measurement model corrupted with noise. The filtered state is reconstructed with a Kalman filter, the chance constraints are reformulated as difference of convex (DC) constraints, and the resulting covariance control problem is reformulated as a DC program, which is solved using successive convexification. The efficiency of the proposed method is illustrated on a double integrator example with varying time horizons, and is compared to other state-of-the-art chance constrained covariance control methods.
comment: v2, submitted to CDC '24
♻ ☆ Coherent Equalization of Linear Quantum Systems
This paper introduces a $H_\infty$-like methodology of coherent filtering for equalization of passive linear quantum systems to help mitigate degrading effects of quantum communication channels. For such systems, which include a wide range of linear quantum optical devices and signals, we seek to find a near optimal equalizing filter which is itself a passive quantum system. The problem amounts to solving an optimization problem subject to constraints dictated by the requirement for the equalizer to be physically realizable. By formulating these constraints in the frequency domain, we show that the problem admits a convex $H_\infty$-like formulation. This allows us to derive a set of suboptimal coherent equalizers using $J$-spectral factorization. An additional semidefinite relaxation combined with the Nevanlinna-Pick interpolation is shown to lead to a tractable algorithm for the design of a near optimal coherent equalizer.
♻ ☆ Forced oscillation source localization from generator measurements
Malfunctioning equipment, erroneous operating conditions or periodic load variations can cause periodic disturbances that would persist over time, creating an undesirable transfer of energy across the system -- an effect referred to as forced oscillations. Wide-area oscillations may damage assets, trigger inadvertent tripping or control actions, and be the cause of equipment failure. Unfortunately, for wide-area oscillations, the location, frequency, and amplitude of these forced oscillations may be hard to determine. Recently, a data-driven maximum-likelihood-based method was proposed to perform source localization in transmission grids under wide-area response scenarios. However, this method relies on full PMU coverage and all buses having inertia and damping. Here, we extend this method to realistic scenarios which includes buses without inertia or dumping, such as passive loads and inverter-based generators. Incorporating Kron reduction directly into the maximum likelihood estimator, we are able to identify the location and frequency of forcing applied at both traditional generators and loads.
comment: 7 pages, 3 figures
♻ ☆ Learning Complex Motion Plans using Neural ODEs with Safety and Stability Guarantees ICRA 2024
We propose a Dynamical System (DS) approach to learn complex, possibly periodic motion plans from kinesthetic demonstrations using Neural Ordinary Differential Equations (NODE). To ensure reactivity and robustness to disturbances, we propose a novel approach that selects a target point at each time step for the robot to follow, by combining tools from control theory and the target trajectory generated by the learned NODE. A correction term to the NODE model is computed online by solving a quadratic program that guarantees stability and safety using control Lyapunov functions and control barrier functions, respectively. Our approach outperforms baseline DS learning techniques on the LASA handwriting dataset and complex periodic trajectories. It is also validated on the Franka Emika robot arm to produce stable motions for wiping and stirring tasks that do not have a single attractor, while being robust to perturbations and safe around humans and obstacles.
comment: accepted to ICRA 2024
♻ ☆ A Traffic Management Framework for On-Demand Urban Air Mobility Systems
Urban Air Mobility (UAM) offers a solution to current traffic congestion by providing on-demand air mobility in urban areas. Effective traffic management is crucial for efficient operation of UAM systems, especially for high-demand scenarios. In this paper, we present a centralized traffic management framework for on-demand UAM systems. Specifically, we provide a scheduling policy, called VertiSync, which schedules the aircraft for either servicing trip requests or rebalancing in the system subject to aircraft safety margins and energy requirements. We characterize the system-level throughput of VertiSync, which determines the demand threshold at which passenger waiting times transition from being stabilized to being increasing over time. We show that the proposed policy is able to maximize throughput for sufficiently large fleet sizes. We demonstrate the performance of VertiSync through a case study for the city of Los Angeles, and show that it significantly reduces passenger waiting times compared to a first-come first-serve scheduling policy.
comment: 9 pages, 6 figures
Robotics 62
☆ ODTFormer: Efficient Obstacle Detection and Tracking with Stereo Cameras Based on Transformer
Obstacle detection and tracking represent a critical component in robot autonomous navigation. In this paper, we propose ODTFormer, a Transformer-based model to address both obstacle detection and tracking problems. For the detection task, our approach leverages deformable attention to construct a 3D cost volume, which is decoded progressively in the form of voxel occupancy grids. We further track the obstacles by matching the voxels between consecutive frames. The entire model can be optimized in an end-to-end manner. Through extensive experiments on DrivingStereo and KITTI benchmarks, our model achieves state-of-the-art performance in the obstacle detection task. We also report comparable accuracy to state-of-the-art obstacle tracking models while requiring only a fraction of their computation cost, typically ten-fold to twenty-fold less. The code and model weights will be publicly released.
comment: 8 pages
☆ SDP Synthesis of Maximum Coverage Trees for Probabilistic Planning under Control Constraints
The paper presents Maximal Covariance Backward Reachable Trees (MAXCOVAR BRT), which is a multi-query algorithm for planning of dynamic systems under stochastic motion uncertainty and constraints on the control input with explicit coverage guarantees. In contrast to existing roadmap-based probabilistic planning methods that sample belief nodes randomly and draw edges between them \cite{csbrm_tro2024}, under control constraints, the reachability of belief nodes needs to be explicitly established and is determined by checking the feasibility of a non-convex program. Moreover, there is no explicit consideration of coverage of the roadmap while adding nodes and edges during the construction procedure for the existing methods. Our contribution is a novel optimization formulation to add nodes and construct the corresponding edge controllers such that the generated roadmap results in provably maximal coverage under control constraints as compared to any other method of adding nodes and edges. We characterize formally the notion of coverage of a roadmap in this stochastic domain via introduction of the h-$\operatorname{BRS}$ (Backward Reachable Set of Distributions) of a tree of distributions under control constraints, and also support our method with extensive simulations on a 6 DoF model.
☆ Extended Reality for Enhanced Human-Robot Collaboration: a Human-in-the-Loop Approach
The rise of automation has provided an opportunity to achieve higher efficiency in manufacturing processes, yet it often compromises the flexibility required to promptly respond to evolving market needs and meet the demand for customization. Human-robot collaboration attempts to tackle these challenges by combining the strength and precision of machines with human ingenuity and perceptual understanding. In this paper, we conceptualize and propose an implementation framework for an autonomous, machine learning-based manipulator that incorporates human-in-the-loop principles and leverages Extended Reality (XR) to facilitate intuitive communication and programming between humans and robots. Furthermore, the conceptual framework foresees human involvement directly in the robot learning process, resulting in higher adaptability and task generalization. The paper highlights key technologies enabling the proposed framework, emphasizing the importance of developing the digital ecosystem as a whole. Additionally, we review the existent implementation approaches of XR in human-robot collaboration, showcasing diverse perspectives and methodologies. The challenges and future outlooks are discussed, delving into the major obstacles and potential research avenues of XR for more natural human-robot interaction and integration in the industrial landscape.
☆ VXP: Voxel-Cross-Pixel Large-scale Image-LiDAR Place Recognition
Recent works on the global place recognition treat the task as a retrieval problem, where an off-the-shelf global descriptor is commonly designed in image-based and LiDAR-based modalities. However, it is non-trivial to perform accurate image-LiDAR global place recognition since extracting consistent and robust global descriptors from different domains (2D images and 3D point clouds) is challenging. To address this issue, we propose a novel Voxel-Cross-Pixel (VXP) approach, which establishes voxel and pixel correspondences in a self-supervised manner and brings them into a shared feature space. Specifically, VXP is trained in a two-stage manner that first explicitly exploits local feature correspondences and enforces similarity of global descriptors. Extensive experiments on the three benchmarks (Oxford RobotCar, ViViD++ and KITTI) demonstrate our method surpasses the state-of-the-art cross-modal retrieval by a large margin.
comment: Project page https://yunjinli.github.io/projects-vxp/
☆ Co-Optimization of Environment and Policies for Decentralized Multi-Agent Navigation
This work views the multi-agent system and its surrounding environment as a co-evolving system, where the behavior of one affects the other. The goal is to take both agent actions and environment configurations as decision variables, and optimize these two components in a coordinated manner to improve some measure of interest. Towards this end, we consider the problem of decentralized multi-agent navigation in cluttered environments. By introducing two sub-objectives of multi-agent navigation and environment optimization, we propose an $\textit{agent-environment co-optimization}$ problem and develop a $\textit{coordinated algorithm}$ that alternates between these sub-objectives to search for an optimal synthesis of agent actions and obstacle configurations in the environment; ultimately, improving the navigation performance. Due to the challenge of explicitly modeling the relation between agents, environment and performance, we leverage policy gradient to formulate a model-free learning mechanism within the coordinated framework. A formal convergence analysis shows that our coordinated algorithm tracks the local minimum trajectory of an associated time-varying non-convex optimization problem. Extensive numerical results corroborate theoretical findings and show the benefits of co-optimization over baselines. Interestingly, the results also indicate that optimized environment configurations are able to offer structural guidance that is key to de-conflicting agents in motion.
☆ Learning Hierarchical Control For Constrained Dynamic Task Assignment
This paper introduces a novel data-driven hierarchical control scheme for managing a fleet of nonlinear, capacity-constrained autonomous agents in an iterative environment. We propose a control framework consisting of a high-level dynamic task assignment and routing layer and low-level motion planning and tracking layer. Each layer of the control hierarchy uses a data-driven MPC policy, maintaining bounded computational complexity at each calculation of a new task assignment or actuation input. We utilize collected data to iteratively refine estimates of agent capacity usage, and update MPC policy parameters accordingly. Our approach leverages tools from iterative learning control to integrate learning at both levels of the hierarchy, and coordinates learning between levels in order to maintain closed-loop feasibility and performance improvement of the connected architecture.
☆ Click to Grasp: Zero-Shot Precise Manipulation via Visual Diffusion Descriptors
Precise manipulation that is generalizable across scenes and objects remains a persistent challenge in robotics. Current approaches for this task heavily depend on having a significant number of training instances to handle objects with pronounced visual and/or geometric part ambiguities. Our work explores the grounding of fine-grained part descriptors for precise manipulation in a zero-shot setting by utilizing web-trained text-to-image diffusion-based generative models. We tackle the problem by framing it as a dense semantic part correspondence task. Our model returns a gripper pose for manipulating a specific part, using as reference a user-defined click from a source image of a visually different instance of the same object. We require no manual grasping demonstrations as we leverage the intrinsic object geometry and features. Practical experiments in a real-world tabletop scenario validate the efficacy of our approach, demonstrating its potential for advancing semantic-aware robotics manipulation. Web page: https://tsagkas.github.io/click2grasp
comment: 8 pages, 4 figures
☆ Physics-Based Causal Reasoning for Safe & Robust Next-Best Action Selection in Robot Manipulation Tasks IROS
Safe and efficient object manipulation is a key enabler of many real-world robot applications. However, this is challenging because robot operation must be robust to a range of sensor and actuator uncertainties. In this paper, we present a physics-informed causal-inference-based framework for a robot to probabilistically reason about candidate actions in a block stacking task in a partially observable setting. We integrate a physics-based simulation of the rigid-body system dynamics with a causal Bayesian network (CBN) formulation to define a causal generative probabilistic model of the robot decision-making process. Using simulation-based Monte Carlo experiments, we demonstrate our framework's ability to successfully: (1) predict block tower stability with high accuracy (Pred Acc: 88.6%); and, (2) select an approximate next-best action for the block stacking task, for execution by an integrated robot system, achieving 94.2% task success rate. We also demonstrate our framework's suitability for real-world robot systems by demonstrating successful task executions with a domestic support robot, with perception and manipulation sub-system integration. Hence, we show that by embedding physics-based causal reasoning into robots' decision-making processes, we can make robot task execution safer, more reliable, and more robust to various types of uncertainty.
comment: 8 pages, 9 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
☆ Bringing Robots Home: The Rise of AI Robots in Consumer Electronics
On March 18, 2024, NVIDIA unveiled Project GR00T, a general-purpose multimodal generative AI model designed specifically for training humanoid robots. Preceding this event, Tesla's unveiling of the Optimus Gen 2 humanoid robot on December 12, 2023, underscored the profound impact robotics is poised to have on reshaping various facets of our daily lives. While robots have long dominated industrial settings, their presence within our homes is a burgeoning phenomenon. This can be attributed, in part, to the complexities of domestic environments and the challenges of creating robots that can seamlessly integrate into our daily routines.
comment: Accepted by IEEE Consumer Electronics Magazine
☆ Exploring 3D Human Pose Estimation and Forecasting from the Robot's Perspective: The HARPER Dataset
We introduce HARPER, a novel dataset for 3D body pose estimation and forecast in dyadic interactions between users and \spot, the quadruped robot manufactured by Boston Dynamics. The key-novelty is the focus on the robot's perspective, i.e., on the data captured by the robot's sensors. These make 3D body pose analysis challenging because being close to the ground captures humans only partially. The scenario underlying HARPER includes 15 actions, of which 10 involve physical contact between the robot and users. The Corpus contains not only the recordings of the built-in stereo cameras of Spot, but also those of a 6-camera OptiTrack system (all recordings are synchronized). This leads to ground-truth skeletal representations with a precision lower than a millimeter. In addition, the Corpus includes reproducible benchmarks on 3D Human Pose Estimation, Human Pose Forecasting, and Collision Prediction, all based on publicly available baseline approaches. This enables future HARPER users to rigorously compare their results with those we provide in this work.
☆ Efficient Model Learning and Adaptive Tracking Control of Magnetic Micro-Robots for Non-Contact Manipulation
Magnetic microrobots can be navigated by an external magnetic field to autonomously move within living organisms with complex and unstructured environments. Potential applications include drug delivery, diagnostics, and therapeutic interventions. Existing techniques commonly impart magnetic properties to the target object,or drive the robot to contact and then manipulate the object, both probably inducing physical damage. This paper considers a non-contact formulation, where the robot spins to generate a repulsive field to push the object without physical contact. Under such a formulation, the main challenge is that the motion model between the input of the magnetic field and the output velocity of the target object is commonly unknown and difficult to analyze. To deal with it, this paper proposes a data-driven-based solution. A neural network is constructed to efficiently estimate the motion model. Then, an approximate model-based optimal control scheme is developed to push the object to track a time-varying trajectory, maintaining the non-contact with distance constraints. Furthermore, a straightforward planner is introduced to assess the adaptability of non-contact manipulation in a cluttered unstructured environment. Experimental results are presented to show the tracking and navigation performance of the proposed scheme.
comment: 7 pages, 6 figures, received by 2024 IEEE International Conference on Robotics and Automation
☆ DaCapo: Accelerating Continuous Learning in Autonomous Systems for Video Analytics
Deep neural network (DNN) video analytics is crucial for autonomous systems such as self-driving vehicles, unmanned aerial vehicles (UAVs), and security robots. However, real-world deployment faces challenges due to their limited computational resources and battery power. To tackle these challenges, continuous learning exploits a lightweight "student" model at deployment (inference), leverages a larger "teacher" model for labeling sampled data (labeling), and continuously retrains the student model to adapt to changing scenarios (retraining). This paper highlights the limitations in state-of-the-art continuous learning systems: (1) they focus on computations for retraining, while overlooking the compute needs for inference and labeling, (2) they rely on power-hungry GPUs, unsuitable for battery-operated autonomous systems, and (3) they are located on a remote centralized server, intended for multi-tenant scenarios, again unsuitable for autonomous systems due to privacy, network availability, and latency concerns. We propose a hardware-algorithm co-designed solution for continuous learning, DaCapo, that enables autonomous systems to perform concurrent executions of inference, labeling, and training in a performant and energy-efficient manner. DaCapo comprises (1) a spatially-partitionable and precision-flexible accelerator enabling parallel execution of kernels on sub-accelerators at their respective precisions, and (2) a spatiotemporal resource allocation algorithm that strategically navigates the resource-accuracy tradeoff space, facilitating optimal decisions for resource allocation to achieve maximal accuracy. Our evaluation shows that DaCapo achieves 6.5% and 5.5% higher accuracy than a state-of-the-art GPU-based continuous learning systems, Ekya and EOMU, respectively, while consuming 254x less power.
☆ A Comparative Study of Real-Time Implementable Cooperative Aerial Manipulation Systems
This survey paper focuses on quadrotor- and multirotor- based cooperative aerial manipulation. Emphasis is first given on comparing and evaluating prototype systems that have been implemented and tested in real-time in diverse application environments. Underlying modeling and control approaches are also discussed and compared. The outcome of the survey allows for understanding the motivation and rationale to develop such systems, their applicability and implementability in diverse applications and also challenges that need to be addressed and overcome. Moreover, the survey provides a guide to develop the next generation of prototype systems based on preferred characteristics, functionality, operability and application domain.
comment: Submitted to MDPI Drones
☆ Tell Me What You Want (What You Really, Really Want): Addressing the Expectation Gap for Goal Conveyance from Humans to Robots
Conveying human goals to autonomous systems (AS) occurs both when the system is being designed and when it is being operated. The design-step conveyance is typically mediated by robotics and AI engineers, who must appropriately capture end-user requirements and concepts of operations, while the operation-step conveyance is mediated by the design, interfaces, and behavior of the AI. However, communication can be difficult during both these periods because of mismatches in the expectations and expertise of the end-user and the roboticist, necessitating more design cycles to resolve. We examine some of the barriers in communicating system design requirements, and develop an augmentation for applied cognitive task analysis (ACTA) methods, that we call robot task analysis (RTA), pertaining specifically to the development of autonomous systems. Further, we introduce a top-down view of an underexplored area of friction between requirements communication -- implied human expectations -- utilizing a collection of work primarily from experimental psychology and social sciences. We show how such expectations can be used in conjunction with task-specific expectations and the system design process for AS to improve design team communication, alleviate barriers to user rejection, and reduce the number of design cycles.
comment: Presented at the End-User Development for Human-Robot Interaction (EUD4HRI) workshop at HRI 2024
☆ Distilling Reinforcement Learning Policies for Interpretable Robot Locomotion: Gradient Boosting Machines and Symbolic Regression
Recent advancements in reinforcement learning (RL) have led to remarkable achievements in robot locomotion capabilities. However, the complexity and ``black-box'' nature of neural network-based RL policies hinder their interpretability and broader acceptance, particularly in applications demanding high levels of safety and reliability. This paper introduces a novel approach to distill neural RL policies into more interpretable forms using Gradient Boosting Machines (GBMs), Explainable Boosting Machines (EBMs) and Symbolic Regression. By leveraging the inherent interpretability of generalized additive models, decision trees, and analytical expressions, we transform opaque neural network policies into more transparent ``glass-box'' models. We train expert neural network policies using RL and subsequently distill them into (i) GBMs, (ii) EBMs, and (iii) symbolic policies. To address the inherent distribution shift challenge of behavioral cloning, we propose to use the Dataset Aggregation (DAgger) algorithm with a curriculum of episode-dependent alternation of actions between expert and distilled policies, to enable efficient distillation of feedback control policies. We evaluate our approach on various robot locomotion gaits -- walking, trotting, bounding, and pacing -- and study the importance of different observations in joint actions for distilled policies using various methods. We train neural expert policies for 205 hours of simulated experience and distill interpretable policies with only 10 minutes of simulated interaction for each gait using the proposed method.
☆ Evaluation and Deployment of LiDAR-based Place Recognition in Dense Forests
Many LiDAR place recognition systems have been developed and tested specifically for urban driving scenarios. Their performance in natural environments such as forests and woodlands have been studied less closely. In this paper, we analyzed the capabilities of four different LiDAR place recognition systems, both handcrafted and learning-based methods, using LiDAR data collected with a handheld device and legged robot within dense forest environments. In particular, we focused on evaluating localization where there is significant translational and orientation difference between corresponding LiDAR scan pairs. This is particularly important for forest survey systems where the sensor or robot does not follow a defined road or path. Extending our analysis we then incorporated the best performing approach, Logg3dNet, into a full 6-DoF pose estimation system -- introducing several verification layers for precise registration. We demonstrated the performance of our methods in three operational modes: online SLAM, offline multi-mission SLAM map merging, and relocalization into a prior map. We evaluated these modes using data captured in forests from three different countries, achieving 80% of correct loop closures candidates with baseline distances up to 5m, and 60% up to 10m.
☆ Exosense: A Vision-Centric Scene Understanding System For Safe Exoskeleton Navigation
Exoskeletons for daily use by those with mobility impairments are being developed. They will require accurate and robust scene understanding systems. Current research has used vision to identify immediate terrain and geometric obstacles, however these approaches are constrained to detections directly in front of the user and are limited to classifying a finite range of terrain types (e.g., stairs, ramps and level-ground). This paper presents Exosense, a vision-centric scene understanding system which is capable of generating rich, globally-consistent elevation maps, incorporating both semantic and terrain traversability information. It features an elastic Atlas mapping framework associated with a visual SLAM pose graph, embedded with open-vocabulary room labels from a Vision-Language Model (VLM). The device's design includes a wide field-of-view (FoV) fisheye multi-camera system to mitigate the challenges introduced by the exoskeleton walking pattern. We demonstrate the system's robustness to the challenges of typical periodic walking gaits, and its ability to construct accurate semantically-rich maps in indoor settings. Additionally, we showcase its potential for motion planning -- providing a step towards safe navigation for exoskeletons.
comment: 8 pages, 10 figures
☆ Bayesian Optimization for Sample-Efficient Policy Improvement in Robotic Manipulation IROS2024
Sample efficient learning of manipulation skills poses a major challenge in robotics. While recent approaches demonstrate impressive advances in the type of task that can be addressed and the sensing modalities that can be incorporated, they still require large amounts of training data. Especially with regard to learning actions on robots in the real world, this poses a major problem due to the high costs associated with both demonstrations and real-world robot interactions. To address this challenge, we introduce BOpt-GMM, a hybrid approach that combines imitation learning with own experience collection. We first learn a skill model as a dynamical system encoded in a Gaussian Mixture Model from a few demonstrations. We then improve this model with Bayesian optimization building on a small number of autonomous skill executions in a sparse reward setting. We demonstrate the sample efficiency of our approach on multiple complex manipulation skills in both simulations and real-world experiments. Furthermore, we make the code and pre-trained models publicly available at http://bopt-gmm. cs.uni-freiburg.de.
comment: 7 pages, 5 figures, 2 tables, submitted to IROS2024
☆ DexDribbler: Learning Dexterous Soccer Manipulation via Dynamic Supervision IROS 2024
Learning dexterous locomotion policy for legged robots is becoming increasingly popular due to its ability to handle diverse terrains and resemble intelligent behaviors. However, joint manipulation of moving objects and locomotion with legs, such as playing soccer, receive scant attention in the learning community, although it is natural for humans and smart animals. A key challenge to solve this multitask problem is to infer the objectives of locomotion from the states and targets of the manipulated objects. The implicit relation between the object states and robot locomotion can be hard to capture directly from the training experience. We propose adding a feedback control block to compute the necessary body-level movement accurately and using the outputs as dynamic joint-level locomotion supervision explicitly. We further utilize an improved ball dynamic model, an extended context-aided estimator, and a comprehensive ball observer to facilitate transferring policy learned in simulation to the real world. We observe that our learning scheme can not only make the policy network converge faster but also enable soccer robots to perform sophisticated maneuvers like sharp cuts and turns on flat surfaces, a capability that was lacking in previous methods. Video and code are available at https://github.com/SysCV/soccer-player
comment: 8 pages, 7 figures, submitted to IROS 2024
☆ Human Reactions to Incorrect Answers from Robots
As robots grow more and more integrated into numerous industries, it is critical to comprehend how humans respond to their failures. This paper systematically studies how trust dynamics and system design are affected by human responses to robot failures. The three-stage survey used in the study provides a thorough understanding of human-robot interactions. While the second stage concentrates on interaction details, such as robot precision and error acknowledgment, the first stage collects demographic data and initial levels of trust. In the last phase, participants' perceptions are examined after the encounter, and trust dynamics, forgiveness, and propensity to suggest robotic technologies are evaluated. Results show that participants' trust in robotic technologies increased significantly when robots acknowledged their errors or limitations to participants and their willingness to suggest robots for activities in the future points to a favorable change in perception, emphasizing the role that direct engagement has in influencing trust dynamics. By providing useful advice for creating more sympathetic, responsive, and reliable robotic systems, the study advances the science of human-robot interaction and promotes a wider adoption of robotic technologies.
comment: 6 pages, 6 figures, 1 table, Ro-Man 2024
☆ UAV-Assisted Maritime Search and Rescue: A Holistic Approach
In this paper, we explore the application of Unmanned Aerial Vehicles (UAVs) in maritime search and rescue (mSAR) missions, focusing on medium-sized fixed-wing drones and quadcopters. We address the challenges and limitations inherent in operating some of the different classes of UAVs, particularly in search operations. Our research includes the development of a comprehensive software framework designed to enhance the efficiency and efficacy of SAR operations. This framework combines preliminary detection onboard UAVs with advanced object detection at ground stations, aiming to reduce visual strain and improve decision-making for operators. It will be made publicly available upon publication. We conduct experiments to evaluate various Region of Interest (RoI) proposal methods, especially by imposing simulated limited bandwidth on them, an important consideration when flying remote or offshore operations. This forces the algorithm to prioritize some predictions over others.
☆ Scene-Graph ViT: End-to-End Open-Vocabulary Visual Relationship Detection
Visual relationship detection aims to identify objects and their relationships in images. Prior methods approach this task by adding separate relationship modules or decoders to existing object detection architectures. This separation increases complexity and hinders end-to-end training, which limits performance. We propose a simple and highly efficient decoder-free architecture for open-vocabulary visual relationship detection. Our model consists of a Transformer-based image encoder that represents objects as tokens and models their relationships implicitly. To extract relationship information, we introduce an attention mechanism that selects object pairs likely to form a relationship. We provide a single-stage recipe to train this model on a mixture of object and relationship detection data. Our approach achieves state-of-the-art relationship detection performance on Visual Genome and on the large-vocabulary GQA benchmark at real-time inference speeds. We provide analyses of zero-shot performance, ablations, and real-world qualitative examples.
☆ ReFeree: Radar-based efficient global descriptor using a Feature and Free space for Place Recognition
Radar is highlighted for robust sensing capabilities in adverse weather conditions (e.g. dense fog, heavy rain, or snowfall). In addition, Radar can cover wide areas and penetrate small particles. Despite these advantages, Radar-based place recognition remains in the early stages compared to other sensors due to its unique characteristics such as low resolution, and significant noise. In this paper, we propose a Radarbased place recognition utilizing a descriptor called ReFeree using a feature and free space. Unlike traditional methods, we overwhelmingly summarize the Radar image. Despite being lightweight, it contains semi-metric information and is also outstanding from the perspective of place recognition performance. For concrete validation, we test a single session from the MulRan dataset and a multi-session from the Oxford Radar RobotCar and the Boreas dataset.
comment: 5 pages, 4 figures
☆ HCTO: Optimality-Aware LiDAR Inertial Odometry with Hybrid Continuous Time Optimization for Compact Wearable Mapping System
Compact wearable mapping system (WMS) has gained significant attention due to their convenience in various applications. Specifically, it provides an efficient way to collect prior maps for 3D structure inspection and robot-based "last-mile delivery" in complex environments. However, vibrations in human motion and the uneven distribution of point cloud features in complex environments often lead to rapid drift, which is a prevalent issue when applying existing LiDAR Inertial Odometry (LIO) methods on low-cost WMS. To address these limitations, we propose a novel LIO for WMSs based on Hybrid Continuous Time Optimization (HCTO) considering the optimality of Lidar correspondences. First, HCTO recognizes patterns in human motion (high-frequency part, low-frequency part, and constant velocity part) by analyzing raw IMU measurements. Second, HCTO constructs hybrid IMU factors according to different motion states, which enables robust and accurate estimation against vibration-induced noise in the IMU measurements. Third, the best point correspondences are selected using optimal design to achieve real-time performance and better odometry accuracy. We conduct experiments on head-mounted WMS datasets to evaluate the performance of our system, demonstrating significant advantages over state-of-the-art methods. Video recordings of experiments can be found on the project page of HCTO: \href{https://github.com/kafeiyin00/HCTO}{https://github.com/kafeiyin00/HCTO}.
☆ Leveraging Large Language Model-based Room-Object Relationships Knowledge for Enhancing Multimodal-Input Object Goal Navigation
Object-goal navigation is a crucial engineering task for the community of embodied navigation; it involves navigating to an instance of a specified object category within unseen environments. Although extensive investigations have been conducted on both end-to-end and modular-based, data-driven approaches, fully enabling an agent to comprehend the environment through perceptual knowledge and perform object-goal navigation as efficiently as humans remains a significant challenge. Recently, large language models have shown potential in this task, thanks to their powerful capabilities for knowledge extraction and integration. In this study, we propose a data-driven, modular-based approach, trained on a dataset that incorporates common-sense knowledge of object-to-room relationships extracted from a large language model. We utilize the multi-channel Swin-Unet architecture to conduct multi-task learning incorporating with multimodal inputs. The results in the Habitat simulator demonstrate that our framework outperforms the baseline by an average of 10.6% in the efficiency metric, Success weighted by Path Length (SPL). The real-world demonstration shows that the proposed approach can efficiently conduct this task by traversing several rooms. For more details and real-world demonstrations, please check our project webpage (https://sunleyuan.github.io/ObjectNav).
comment: will soon submit to the Elsevier journal, Advanced Engineering Informatics
☆ Extrinsic Calibration of Multiple LiDARs for a Mobile Robot based on Floor Plane And Object Segmentation
Mobile robots equipped with multiple light detection and ranging (LiDARs) and capable of recognizing their surroundings are increasing due to the minitualization and cost reduction of LiDAR. This paper proposes a target-less extrinsic calibration method of multiple LiDARs with non-overlapping field of view (FoV). The proposed method uses accumulated point clouds of floor plane and objects while in motion. It enables accurate calibration with challenging configuration of LiDARs that directed towards the floor plane, caused by biased feature values. Additionally, the method includes a noise removal module that considers the scanning pattern to address bleeding points, which are noises of significant source of error in point cloud alignment using high-density LiDARs. Evaluations through simulation demonstrate that the proposed method achieved higher accuracy extrinsic calibration with two and four LiDARs than conventional methods, regardless type of objects. Furthermore, the experiments using a real mobile robot has shown that our proposed noise removal module can eliminate noise more precisely than conventional methods, and the estimated extrinsic parameters have successfully created consistent 3D maps.
comment: 8pages, 10figures
☆ Development of a Compact Robust Passive Transformable Omni-Ball for Enhanced Step-Climbing and Vibration Reduction
This paper introduces the Passive Transformable Omni-Ball (PTOB), an advanced omnidirectional wheel engineered to enhance step-climbing performance, incorporate built-in actuators, diminish vibrations, and fortify structural integrity. By modifying the omni-ball's structure from two to three segments, we have achieved improved in-wheel actuation and a reduction in vibrational feedback. Additionally, we have implemented a sliding mechanism in the follower wheels to boost the wheel's step-climbing abilities. A prototype with a 127 mm diameter PTOB was constructed, which confirmed its functionality for omnidirectional movement and internal actuation. Compared to a traditional omni-wheel, the PTOB demonstrated a comparable level of vibration while offering superior capabilities. Extensive testing in varied settings showed that the PTOB can adeptly handle step obstacles up to 45 mm, equivalent to 35 $\%$ of the wheel's diameter, in both the forward and lateral directions. The PTOB showcased robust construction and proved to be versatile in navigating through environments with diverse obstacles.
comment: 8 pages, 16 figures
☆ Robust Locomotion via Zero-order Stochastic Nonlinear Model Predictive Control with Guard Saltation Matrix
This paper presents a stochastic/robust nonlinear model predictive control (NMPC) to enhance the robustness of legged locomotion against contact uncertainties. We integrate the contact uncertainties into the covariance propagation of stochastic/robust NMPC framework by leveraging the guard saltation matrix and an extended Kalman filter-like covariance update. We achieve fast stochastic/robust NMPC computation by utilizing the zero-order stochastic/robust NMPC algorithm with additional improvements in computational efficiency concerning the feedback gains. We conducted numerical experiments and demonstrate that the proposed method can accurately forecast future state covariance and generate trajectories that satisfies constraints even in the presence of the contact uncertainties. Hardware experiments on the perceptive locomotion of a wheeled-legged robot were also carried out, validating the feasibility of the proposed method in a real-world system with limited on-board computation.
comment: 8 pages, 8 figures
☆ Evidential Semantic Mapping in Off-road Environments with Uncertainty-aware Bayesian Kernel Inference
Robotic mapping with Bayesian Kernel Inference (BKI) has shown promise in creating semantic maps by effectively leveraging local spatial information. However, existing semantic mapping methods face challenges in constructing reliable maps in unstructured outdoor scenarios due to unreliable semantic predictions. To address this issue, we propose an evidential semantic mapping, which can enhance reliability in perceptually challenging off-road environments. We integrate Evidential Deep Learning into the semantic segmentation network to obtain the uncertainty estimate of semantic prediction. Subsequently, this semantic uncertainty is incorporated into an uncertainty-aware BKI, tailored to prioritize more confident semantic predictions when accumulating semantic information. By adaptively handling semantic uncertainties, the proposed framework constructs robust representations of the surroundings even in previously unseen environments. Comprehensive experiments across various off-road datasets demonstrate that our framework enhances accuracy and robustness, consistently outperforming existing methods in scenes with high perceptual uncertainties.
comment: Our project website can be found at https://kjyoung.github.io/Homepage/#/Projects/Evidential-Semantic-Mapping
☆ Semantics from Space: Satellite-Guided Thermal Semantic Segmentation Annotation for Aerial Field Robots
We present a new method to automatically generate semantic segmentation annotations for thermal imagery captured from an aerial vehicle by utilizing satellite-derived data products alongside onboard global positioning and attitude estimates. This new capability overcomes the challenge of developing thermal semantic perception algorithms for field robots due to the lack of annotated thermal field datasets and the time and costs of manual annotation, enabling precise and rapid annotation of thermal data from field collection efforts at a massively-parallelizable scale. By incorporating a thermal-conditioned refinement step with visual foundation models, our approach can produce highly-precise semantic segmentation labels using low-resolution satellite land cover data for little-to-no cost. It achieves 98.5% of the performance from using costly high-resolution options and demonstrates between 70-160% improvement over popular zero-shot semantic segmentation methods based on large vision-language models currently used for generating annotations for RGB imagery. Code will be available at: https://github.com/connorlee77/aerial-auto-segment.
☆ A Roadmap Towards Automated and Regulated Robotic Systems
The rapid development of generative technology opens up possibility for higher level of automation, and artificial intelligence (AI) embodiment in robotic systems is imminent. However, due to the blackbox nature of the generative technology, the generation of the knowledge and workflow scheme is uncontrolled, especially in a dynamic environment and a complex scene. This poses challenges to regulations in safety-demanding applications such as medical scenes. We argue that the unregulated generative processes from AI is fitted for low level end tasks, but intervention in the form of manual or automated regulation should happen post-workflow-generation and pre-robotic-execution. To address this, we propose a roadmap that can lead to fully automated and regulated robotic systems. In this paradigm, the high level policies are generated as structured graph data, enabling regulatory oversight and reusability, while the code base for lower level tasks is generated by generative models. Our approach aims the transitioning from expert knowledge to regulated action, akin to the iterative processes of study, practice, scrutiny, and execution in human tasks. We identify the generative and deterministic processes in a design cycle, where generative processes serve as a text-based world simulator and the deterministic processes generate the executable system. We propose State Machine Seralization Language (SMSL) to be the conversion point between text simulator and executable workflow control. From there, we analyze the modules involved based on the current literature, and discuss human in the loop. As a roadmap, this work identifies the current possible implementation and future work. This work does not provide an implemented system but envisions to inspire the researchers working on the direction in the roadmap. We implement the SMSL and D-SFO paradigm that serve as the starting point of the roadmap.
comment: 17 pages, 9 figures
☆ GelLink: A Compact Multi-phalanx Finger with Vision-based Tactile Sensing and Proprioception ICRA 2024
Compared to fully-actuated robotic end-effectors, underactuated ones are generally more adaptive, robust, and cost-effective. However, state estimation for underactuated hands is usually more challenging. Vision-based tactile sensors, like Gelsight, can mitigate this issue by providing high-resolution tactile sensing and accurate proprioceptive sensing. As such, we present GelLink, a compact, underactuated, linkage-driven robotic finger with low-cost, high-resolution vision-based tactile sensing and proprioceptive sensing capabilities. In order to reduce the amount of embedded hardware, i.e. the cameras and motors, we optimize the linkage transmission with a planar linkage mechanism simulator and develop a planar reflection simulator to simplify the tactile sensing hardware. As a result, GelLink only requires one motor to actuate the three phalanges, and one camera to capture tactile signals along the entire finger. Overall, GelLink is a compact robotic finger that shows adaptability and robustness when performing grasping tasks. The integration of vision-based tactile sensors can significantly enhance the capabilities of underactuated fingers and potentially broaden their future usage.
comment: Supplement video: https://www.youtube.com/watch?v=hZwUpAig5C0 . 7 pages, 9 figures. ICRA 2024 (IEEE International Conference on Robotics and Automation)
☆ Learning to Change: Choreographing Mixed Traffic Through Lateral Control and Hierarchical Reinforcement Learning
The management of mixed traffic that consists of robot vehicles (RVs) and human-driven vehicles (HVs) at complex intersections presents a multifaceted challenge. Traditional signal controls often struggle to adapt to dynamic traffic conditions and heterogeneous vehicle types. Recent advancements have turned to strategies based on reinforcement learning (RL), leveraging its model-free nature, real-time operation, and generalizability over different scenarios. We introduce a hierarchical RL framework to manage mixed traffic through precise longitudinal and lateral control of RVs. Our proposed hierarchical framework combines the state-of-the-art mixed traffic control algorithm as a high level decision maker to improve the performance and robustness of the whole system. Our experiments demonstrate that the framework can reduce the average waiting time by up to 54% compared to the state-of-the-art mixed traffic control method. When the RV penetration rate exceeds 60%, our technique consistently outperforms conventional traffic signal control programs in terms of the average waiting time for all vehicles at the intersection.
☆ TEeVTOL: Balancing Energy and Time Efficiency in eVTOL Aircraft Path Planning Across City-Scale Wind Fields
Electric vertical-takeoff and landing (eVTOL) aircraft, recognized for their maneuverability and flexibility, offer a promising alternative to our transportation system. However, the operational effectiveness of these aircraft faces many challenges, such as the delicate balance between energy and time efficiency, stemming from unpredictable environmental factors, including wind fields. Mathematical modeling-based approaches have been adopted to plan aircraft flight path in urban wind fields with the goal to save energy and time costs. While effective, they are limited in adapting to dynamic and complex environments. To optimize energy and time efficiency in eVTOL's flight through dynamic wind fields, we introduce a novel path planning method leveraging deep reinforcement learning. We assess our method with extensive experiments, comparing it to Dijkstra's algorithm -- the theoretically optimal approach for determining shortest paths in a weighted graph, where weights represent either energy or time cost. The results show that our method achieves a graceful balance between energy and time efficiency, closely resembling the theoretically optimal values for both objectives.
Learning Quadruped Locomotion Using Differentiable Simulation
While most recent advancements in legged robot control have been driven by model-free reinforcement learning, we explore the potential of differentiable simulation. Differentiable simulation promises faster convergence and more stable training by computing low-variant first-order gradients using the robot model, but so far, its use for legged robot control has remained limited to simulation. The main challenge with differentiable simulation lies in the complex optimization landscape of robotic tasks due to discontinuities in contact-rich environments, e.g., quadruped locomotion. This work proposes a new, differentiable simulation framework to overcome these challenges. The key idea involves decoupling the complex whole-body simulation, which may exhibit discontinuities due to contact, into two separate continuous domains. Subsequently, we align the robot state resulting from the simplified model with a more precise, non-differentiable simulator to maintain sufficient simulation accuracy. Our framework enables learning quadruped walking in minutes using a single simulated robot without any parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills, including trot, pace, bound, and gallop, on challenging terrains in minutes. Additionally, our policy achieves robust locomotion performance in the real world zero-shot. To the best of our knowledge, this work represents the first demonstration of using differentiable simulation for controlling a real quadruped robot. This work provides several important insights into using differentiable simulations for legged locomotion in the real world.
☆ Multi-agent Task-Driven Exploration via Intelligent Map Compression and Sharing
This paper investigates the task-driven exploration of unknown environments with mobile sensors communicating compressed measurements. The sensors explore the area and transmit their compressed data to another robot, assisting it in reaching a goal location. We propose a novel communication framework and a tractable multi-agent exploration algorithm to select the sensors' actions. The algorithm uses a task-driven measure of uncertainty, resulting from map compression, as a reward function. We validate the efficacy of our algorithm through numerical simulations conducted on a realistic map and compare it with two alternative approaches. The results indicate that the proposed algorithm effectively decreases the time required for the robot to reach its target without causing excessive load on the communication network.
☆ Multiple and Gyro-Free Inertial Datasets
An inertial navigation system (INS) utilizes three orthogonal accelerometers and gyroscopes to determine platform position, velocity, and orientation. There are countless applications for INS, including robotics, autonomous platforms, and the internet of things. Recent research explores the integration of data-driven methods with INS, highlighting significant innovations, improving accuracy and efficiency. Despite the growing interest in this field and the availability of INS datasets, no datasets are available for gyro-free INS (GFINS) and multiple inertial measurement unit (MIMU) architectures. To fill this gap and to stimulate further research in this field, we designed and recorded GFINS and MIMU datasets using 54 inertial sensors grouped in nine inertial measurement units. These sensors can be used to define and evaluate different types of MIMU and GFINS architectures. The inertial sensors were arranged in three different sensor configurations and mounted on a mobile robot and a passenger car. In total, the dataset contains 35 hours of inertial data and corresponding ground truth trajectories. The data and code are freely accessible through our GitHub repository.
comment: 10 pages, 16 figures, 6 tables
♻ ☆ A Modular Aerial System Based on Homogeneous Quadrotors with Fault-Tolerant Control ICRA2024
The standard quadrotor is one of the most popular and widely used aerial vehicle of recent decades, offering great maneuverability with mechanical simplicity. However, the under-actuation characteristic limits its applications, especially when it comes to generating desired wrench with six degrees of freedom (DOF). Therefore, existing work often compromises between mechanical complexity and the controllable DOF of the aerial system. To take advantage of the mechanical simplicity of a standard quadrotor, we propose a modular aerial system, IdentiQuad, that combines only homogeneous quadrotor-based modules. Each IdentiQuad can be operated alone like a standard quadrotor, but at the same time allows task-specific assembly, increasing the controllable DOF of the system. Each module is interchangeable within its assembly. We also propose a general controller for different configurations of assemblies, capable of tolerating rotor failures and balancing the energy consumption of each module. The functionality and robustness of the system and its controller are validated using physics-based simulations for different assembly configurations.
comment: ICRA2024
♻ ☆ Instance-aware Exploration-Verification-Exploitation for Instance ImageGoal Navigation
As a new embodied vision task, Instance ImageGoal Navigation (IIN) aims to navigate to a specified object depicted by a goal image in an unexplored environment. The main challenge of this task lies in identifying the target object from different viewpoints while rejecting similar distractors. Existing ImageGoal Navigation methods usually adopt the simple Exploration-Exploitation framework and ignore the identification of specific instance during navigation. In this work, we propose to imitate the human behaviour of ``getting closer to confirm" when distinguishing objects from a distance. Specifically, we design a new modular navigation framework named Instance-aware Exploration-Verification-Exploitation (IEVE) for instance-level image goal navigation. Our method allows for active switching among the exploration, verification, and exploitation actions, thereby facilitating the agent in making reasonable decisions under different situations. On the challenging HabitatMatterport 3D semantic (HM3D-SEM) dataset, our method surpasses previous state-of-the-art work, with a classical segmentation model (0.684 vs. 0.561 success) or a robust model (0.702 vs. 0.561 success). Our code will be made publicly available at https://github.com/XiaohanLei/IEVE.
♻ ☆ Learning a Depth Covariance Function CVPR 2023
We propose learning a depth covariance function with applications to geometric vision tasks. Given RGB images as input, the covariance function can be flexibly used to define priors over depth functions, predictive distributions given observations, and methods for active point selection. We leverage these techniques for a selection of downstream tasks: depth completion, bundle adjustment, and monocular dense visual odometry.
comment: CVPR 2023. Project page: https://edexheim.github.io/DepthCov/
♻ ☆ Deep learning reduces sensor requirements for gust rejection on a small uncrewed aerial vehicle morphing wing
There is a growing need for uncrewed aerial vehicles (UAVs) to operate in cities. However, the uneven urban landscape and complex street systems cause large-scale wind gusts that challenge the safe and effective operation of UAVs. Current gust alleviation methods rely on traditional control surfaces and computationally expensive modeling to select a control action, leading to a slower response. Here, we used deep reinforcement learning to create an autonomous gust alleviation controller for a camber-morphing wing. This method reduced gust impact by 84%, directly from real-time, on-board pressure signals. Notably, we found that gust alleviation using signals from only three pressure taps was statistically indistinguishable from using six signals. This reduced-sensor fly-by-feel control opens the door to UAV missions in previously inoperable locations.
♻ ☆ Exploring Human's Gender Perception and Bias toward Non-Humanoid Robots
As non-humanoid robots increasingly permeate various sectors, understanding their design implications for human acceptance becomes paramount. Despite their ubiquity, studies on how to improve human interaction are sparse. Our investigation, conducted through two surveys, addresses this gap. The first survey emphasizes non-humanoid robots and human perceptions about gender attributions, suggesting that both design and perceived gender influence acceptance. Survey 2 investigates the effects of varying gender cues on robot designs and their consequent impacts on human-robot interactions. Our findings highlighted that distinct gender cues can bolster or impede interaction comfort.
♻ ☆ Star-Searcher: A Complete and Efficient Aerial System for Autonomous Target Search in Complex Unknown Environments
This paper tackles the challenge of autonomous target search using unmanned aerial vehicles (UAVs) in complex unknown environments. To fill the gap in systematic approaches for this task, we introduce Star-Searcher, an aerial system featuring specialized sensor suites, mapping, and planning modules to optimize searching. Path planning challenges due to increased inspection requirements are addressed through a hierarchical planner with a visibility-based viewpoint clustering method. This simplifies planning by breaking it into global and local sub-problems, ensuring efficient global and local path coverage in real-time. Furthermore, our global path planning employs a history-aware mechanism to reduce motion inconsistency from frequent map changes, significantly enhancing search efficiency. We conduct comparisons with state-of-the-art methods in both simulation and the real world, demonstrating shorter flight paths, reduced time, and higher target search completeness. Our approach will be open-sourced for community benefit at https://github.com/SYSU-STAR/STAR-Searcher.
comment: Aceepted to IEEE RA-L. Code: https://github.com/SYSU-STAR/STAR-Searcher. Video: https://www.youtube.com/watch?v=08ll_oo_DtU
♻ ☆ Large Language Models for Multi-Modal Human-Robot Interaction
This paper presents an innovative large language model (LLM)-based robotic system for enhancing multi-modal human-robot interaction (HRI). Traditional HRI systems relied on complex designs for intent estimation, reasoning, and behavior generation, which were resource-intensive. In contrast, our system empowers researchers and practitioners to regulate robot behavior through three key aspects: providing high-level linguistic guidance, creating "atomics" for actions and expressions the robot can use, and offering a set of examples. Implemented on a physical robot, it demonstrates proficiency in adapting to multi-modal inputs and determining the appropriate manner of action to assist humans with its arms, following researchers' defined guidelines. Simultaneously, it coordinates the robot's lid, neck, and ear movements with speech output to produce dynamic, multi-modal expressions. This showcases the system's potential to revolutionize HRI by shifting from conventional, manual state-and-flow design methods to an intuitive, guidance-based, and example-driven approach.
comment: 10 pages, 6 figures
♻ ☆ MAkEable: Memory-centered and Affordance-based Task Execution Framework for Transferable Mobile Manipulation Skills
To perform versatile mobile manipulation tasks in human-centered environments, the ability to efficiently transfer learned tasks and experiences from one robot to another or across different environments is key. In this paper, we present MAkEable, a versatile uni- and multi-manual mobile manipulation framework that facilitates the transfer of capabilities and knowledge across different tasks, environments, and robots. Our framework integrates an affordance-based task description into the memory-centric cognitive architecture of the ARMAR humanoid robot family, which supports the sharing of experiences and demonstrations for transfer learning. By representing mobile manipulation actions through affordances, i.e., interaction possibilities of the robot with its environment, we provide a unifying framework for the autonomous uni- and multi-manual manipulation of known and unknown objects in various environments. We demonstrate the applicability of the framework in real-world experiments for multiple robots, tasks, and environments. This includes grasping known and unknown objects, object placing, bimanual object grasping, memory-enabled skill transfer in a drawer opening scenario across two different humanoid robots, and a pouring task learned from human demonstration.
♻ ☆ Driving Animatronic Robot Facial Expression From Speech
Animatronic robots aim to enable natural human-robot interaction through lifelike facial expressions. However, generating realistic, speech-synchronized robot expressions is challenging due to the complexities of facial biomechanics and responsive motion synthesis. This paper presents a principled, skinning-centric approach to drive animatronic robot facial expressions from speech. The proposed approach employs linear blend skinning (LBS) as the core representation to guide tightly integrated innovations in embodiment design and motion synthesis. LBS informs the actuation topology, enables human expression retargeting, and allows speech-driven facial motion generation. The proposed approach is capable of generating highly realistic, real-time facial expressions from speech on an animatronic face, significantly advancing robots' ability to replicate nuanced human expressions for natural interaction.
comment: Under review. For associated project page, see https://library87.github.io/animatronic-face-iros24
♻ ☆ ROS-Causal: A ROS-based Causal Analysis Framework for Human-Robot Interaction Applications
Deploying robots in human-shared spaces requires understanding interactions among nearby agents and objects. Modelling cause-and-effect relations through causal inference aids in predicting human behaviours and anticipating robot interventions. However, a critical challenge arises as existing causal discovery methods currently lack an implementation inside the ROS ecosystem, the standard de facto in robotics, hindering effective utilisation in robotics. To address this gap, this paper introduces ROS-Causal, a ROS-based framework for onboard data collection and causal discovery in human-robot spatial interactions. An ad-hoc simulator, integrated with ROS, illustrates the approach's effectiveness, showcasing the robot onboard generation of causal models during data collection. ROS-Causal is available on GitHub: https://github.com/lcastri/roscausal.git.
comment: Accepted by the "Causal-HRI: Causal Learning for Human-Robot Interaction" workshop at the 2024 ACM/IEEE International Conference on Human-Robot Interaction (HRI)
♻ ☆ Exploring Large Language Models to Facilitate Variable Autonomy for Human-Robot Teaming
In a rapidly evolving digital landscape autonomous tools and robots are becoming commonplace. Recognizing the significance of this development, this paper explores the integration of Large Language Models (LLMs) like Generative pre-trained transformer (GPT) into human-robot teaming environments to facilitate variable autonomy through the means of verbal human-robot communication. In this paper, we introduce a novel framework for such a GPT-powered multi-robot testbed environment, based on a Unity Virtual Reality (VR) setting. This system allows users to interact with robot agents through natural language, each powered by individual GPT cores. By means of OpenAI's function calling, we bridge the gap between unstructured natural language input and structure robot actions. A user study with 12 participants explores the effectiveness of GPT-4 and, more importantly, user strategies when being given the opportunity to converse in natural language within a multi-robot environment. Our findings suggest that users may have preconceived expectations on how to converse with robots and seldom try to explore the actual language and cognitive capabilities of their robot collaborators. Still, those users who did explore where able to benefit from a much more natural flow of communication and human-like back-and-forth. We provide a set of lessons learned for future research and technical implementations of similar systems.
comment: Frontiers in Robotics and AI, Variable Autonomy for Human-Robot Teaming
♻ ☆ SLIM: Skill Learning with Multiple Critics ICRA 2024
Self-supervised skill learning aims to acquire useful behaviors that leverage the underlying dynamics of the environment. Latent variable models, based on mutual information maximization, have been successful in this task but still struggle in the context of robotic manipulation. As it requires impacting a possibly large set of degrees of freedom composing the environment, mutual information maximization fails alone in producing useful and safe manipulation behaviors. Furthermore, tackling this by augmenting skill discovery rewards with additional rewards through a naive combination might fail to produce desired behaviors. To address this limitation, we introduce SLIM, a multi-critic learning approach for skill discovery with a particular focus on robotic manipulation. Our main insight is that utilizing multiple critics in an actor-critic framework to gracefully combine multiple reward functions leads to a significant improvement in latent-variable skill discovery for robotic manipulation while overcoming possible interference occurring among rewards which hinders convergence to useful skills. Furthermore, in the context of tabletop manipulation, we demonstrate the applicability of our novel skill discovery approach to acquire safe and efficient motor primitives in a hierarchical reinforcement learning fashion and leverage them through planning, significantly surpassing baseline approaches for skill discovery.
comment: Accepted at IEEE ICRA 2024
♻ ☆ R2SNet: Scalable Domain Adaptation for Object Detection in Cloud-Based Robots Ecosystems via Proposal Refinement
We introduce a novel approach for scalable domain adaptation in cloud robotics scenarios where robots rely on third-party AI inference services powered by large pre-trained deep neural networks. Our method is based on a downstream proposal-refinement stage running locally on the robots, exploiting a new lightweight DNN architecture, R2SNet. This architecture aims to mitigate performance degradation from domain shifts by adapting the object detection process to the target environment, focusing on relabeling, rescoring, and suppression of bounding-box proposals. Our method allows for local execution on robots, addressing the scalability challenges of domain adaptation without incurring significant computational costs. Real-world results on mobile service robots performing door detection show the effectiveness of the proposed method in achieving scalable domain adaptation.
♻ ☆ CoBRA: A Composable Benchmark for Robotics Applications ICRA'24
Selecting an optimal robot, its base pose, and trajectory for a given task is currently mainly done by human expertise or trial and error. To evaluate automatic approaches to this combined optimization problem, we introduce a benchmark suite encompassing a unified format for robots, environments, and task descriptions. Our benchmark suite is especially useful for modular robots, where the multitude of robots that can be assembled creates a host of additional parameters to optimize. We include tasks such as machine tending and welding in synthetic environments and 3D scans of real-world machine shops. All benchmarks are accessible through https://cobra.cps.cit.tum.de, a platform to conveniently share, reference, and compare tasks, robot models, and solutions.
comment: 7 pages, 5 Figures, 5 Tables Final version for IEEE ICRA'24
♻ ☆ Generalized Early Stopping in Evolutionary Direct Policy Search
Lengthy evaluation times are common in many optimization problems such as direct policy search tasks, especially when they involve conducting evaluations in the physical world, e.g. in robotics applications. Often when evaluating solution over a fixed time period it becomes clear that the objective value will not increase with additional computation time (for example when a two wheeled robot continuously spins on the spot). In such cases, it makes sense to stop the evaluation early to save computation time. However, most approaches to stop the evaluation are problem specific and need to be specifically designed for the task at hand. Therefore, we propose an early stopping method for direct policy search. The proposed method only looks at the objective value at each time step and requires no problem specific knowledge. We test the introduced stopping criterion in five direct policy search environments drawn from games, robotics and classic control domains, and show that it can save up to 75% of the computation time. We also compare it with problem specific stopping criteria and show that it performs comparably, while being more generally applicable.
♻ ☆ Language and Sketching: An LLM-driven Interactive Multimodal Multitask Robot Navigation Framework
The socially-aware navigation system has evolved to adeptly avoid various obstacles while performing multiple tasks, such as point-to-point navigation, human-following, and -guiding. However, a prominent gap persists: in Human-Robot Interaction (HRI), the procedure of communicating commands to robots demands intricate mathematical formulations. Furthermore, the transition between tasks does not quite possess the intuitive control and user-centric interactivity that one would desire. In this work, we propose an LLM-driven interactive multimodal multitask robot navigation framework, termed LIM2N, to solve the above new challenge in the navigation field. We achieve this by first introducing a multimodal interaction framework where language and hand-drawn inputs can serve as navigation constraints and control objectives. Next, a reinforcement learning agent is built to handle multiple tasks with the received information. Crucially, LIM2N creates smooth cooperation among the reasoning of multimodal input, multitask planning, and adaptation and processing of the intelligent sensing modules in the complicated system. Extensive experiments are conducted in both simulation and the real world demonstrating that LIM2N has superior user needs understanding, alongside an enhanced interactive experience.
♻ ☆ Real-time Perceptive Motion Control using Control Barrier Functions with Analytical Smoothing for Six-Wheeled-Telescopic-Legged Robot Tachyon 3
To achieve safe legged locomotion, it is important to generate motion in real-time considering various constraints in robots and environments. In this study, we propose a lightweight real-time perspective motion control system for the newly developed six-wheeled-telescopic-legged robot, Tachyon 3. In the proposed method, analytically smoothed constraints including Smooth Separating Axis Theorem (Smooth SAT) as a novel higher order differentiable collision detection for 3D shapes is applied to the Control Barrier Function (CBF). The proposed system integrating the CBF achieves online motion generation in a short control cycle of 1 ms that satisfies joint limitations, environmental collision avoidance and safe convex foothold constraints. The efficiency of Smooth SAT is shown from the collision detection time of 1 us or less and the CBF constraint computation time for Tachyon3 of several us. Furthermore, the effectiveness of the proposed system is verified through the stair-climbing motion, integrating online recognition in a simulation and a real machine.
comment: 8 pages, 8 figures, This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Distilling and Retrieving Generalizable Knowledge for Robot Manipulation via Language Corrections
Today's robot policies exhibit subpar performance when faced with the challenge of generalizing to novel environments. Human corrective feedback is a crucial form of guidance to enable such generalization. However, adapting to and learning from online human corrections is a non-trivial endeavor: not only do robots need to remember human feedback over time to retrieve the right information in new settings and reduce the intervention rate, but also they would need to be able to respond to feedback that can be arbitrary corrections about high-level human preferences to low-level adjustments to skill parameters. In this work, we present Distillation and Retrieval of Online Corrections (DROC), a large language model (LLM)-based system that can respond to arbitrary forms of language feedback, distill generalizable knowledge from corrections, and retrieve relevant past experiences based on textual and visual similarity for improving performance in novel settings. DROC is able to respond to a sequence of online language corrections that address failures in both high-level task plans and low-level skill primitives. We demonstrate that DROC effectively distills the relevant information from the sequence of online corrections in a knowledge base and retrieves that knowledge in settings with new task or object instances. DROC outperforms other techniques that directly generate robot code via LLMs by using only half of the total number of corrections needed in the first round and requires little to no corrections after two iterations. We show further results, videos, prompts and code on https://sites.google.com/stanford.edu/droc .
comment: 8 pages, 4 figures, videos and code links on website https://sites.google.com/stanford.edu/droc
♻ ☆ Deep Learning for Inertial Positioning: A Survey
Inertial sensors are widely utilized in smartphones, drones, robots, and IoT devices, playing a crucial role in enabling ubiquitous and reliable localization. Inertial sensor-based positioning is essential in various applications, including personal navigation, location-based security, and human-device interaction. However, low-cost MEMS inertial sensors' measurements are inevitably corrupted by various error sources, leading to unbounded drifts when integrated doubly in traditional inertial navigation algorithms, subjecting inertial positioning to the problem of error drifts. In recent years, with the rapid increase in sensor data and computational power, deep learning techniques have been developed, sparking significant research into addressing the problem of inertial positioning. Relevant literature in this field spans across mobile computing, robotics, and machine learning. In this article, we provide a comprehensive review of deep learning-based inertial positioning and its applications in tracking pedestrians, drones, vehicles, and robots. We connect efforts from different fields and discuss how deep learning can be applied to address issues such as sensor calibration, positioning error drift reduction, and multi-sensor fusion. This article aims to attract readers from various backgrounds, including researchers and practitioners interested in the potential of deep learning-based techniques to solve inertial positioning problems. Our review demonstrates the exciting possibilities that deep learning brings to the table and provides a roadmap for future research in this field.
comment: Accepted by IEEE Transactions on Intelligent Transportation Systems
♻ ☆ AMP: Autoregressive Motion Prediction Revisited with Next Token Prediction for Autonomous Driving
As an essential task in autonomous driving (AD), motion prediction aims to predict the future states of surround objects for navigation. One natural solution is to estimate the position of other agents in a step-by-step manner where each predicted time-step is conditioned on both observed time-steps and previously predicted time-steps, i.e., autoregressive prediction. Pioneering works like SocialLSTM and MFP design their decoders based on this intuition. However, almost all state-of-the-art works assume that all predicted time-steps are independent conditioned on observed time-steps, where they use a single linear layer to generate positions of all time-steps simultaneously. They dominate most motion prediction leaderboards due to the simplicity of training MLPs compared to autoregressive networks. In this paper, we introduce the GPT style next token prediction into motion forecasting. In this way, the input and output could be represented in a unified space and thus the autoregressive prediction becomes more feasible. However, different from language data which is composed of homogeneous units -words, the elements in the driving scene could have complex spatial-temporal and semantic relations. To this end, we propose to adopt three factorized attention modules with different neighbors for information aggregation and different position encoding styles to capture their relations, e.g., encoding the transformation between coordinate systems for spatial relativity while adopting RoPE for temporal relativity. Empirically, by equipping with the aforementioned tailored designs, the proposed method achieves state-of-the-art performance in the Waymo Open Motion and Waymo Interaction datasets. Notably, AMP outperforms other recent autoregressive motion prediction methods: MotionLM and StateTransformer, which demonstrates the effectiveness of the proposed designs.
♻ ☆ Redundancy parameterization and inverse kinematics of 7-DOF revolute manipulators
Seven degree-of-freedom (DOF) robot arms have one redundant DOF which does not change the motion of the end effector. The redundant DOF offers greater manipulability of the arm configuration to avoid obstacles and singularities, but it must be parameterized to fully specify the joint angles for a given end effector pose. For 7-DOF revolute (7R) manipulators, we introduce a new concept of generalized shoulder-elbow-wrist (SEW) angle, a generalization of the conventional SEW angle but with an arbitrary choice of the reference direction function. The SEW angle is widely used and easy for human operators to visualize as a rotation of the elbow about the shoulder-wrist line. Since other redundancy parameterizations including the conventional SEW angle encounter an algorithmic singularity along a line in the workspace, we introduce a special choice of the reference direction function called the stereographic SEW angle which has a singularity only along a half-line, which can be placed out of reach. We prove that such a singularity is unavoidable for any parameterization. We also include expressions for the SEW angle Jacobian along with singularity analysis. Finally, we provide efficient and singularity-robust inverse kinematics solutions for most known 7R manipulators using the general SEW angle and the subproblem decomposition method. These solutions are often closed-form but may sometimes involve a 1D or 2D search in the general case. Search-based solutions may be converted to finding zeros of a high-order polynomial. Inverse kinematics solutions, examples, and evaluations are available in a publicly accessible repository.
comment: 22 pages, 14 figures. Update: Sawyer IK using polynomial method, two video extensions, expanded related literature
Open X-Embodiment: Robotic Learning Datasets and RT-X Models
Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with general pretrained backbones serving as a starting point for many applications. Can such a consolidation happen in robotics? Conventionally, robotic learning methods train a separate model for every application, every robot, and even every environment. Can we instead train generalist X-robot policy that can be adapted efficiently to new robots, tasks, and environments? In this paper, we provide datasets in standardized data formats and models to make it possible to explore this possibility in the context of robotic manipulation, alongside experimental results that provide an example of effective X-robot policies. We assemble a dataset from 22 different robots collected through a collaboration between 21 institutions, demonstrating 527 skills (160266 tasks). We show that a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves the capabilities of multiple robots by leveraging experience from other platforms. More details can be found on the project website https://robotics-transformer-x.github.io.
comment: Project website: https://robotics-transformer-x.github.io
TD-MPC2: Scalable, Robust World Models for Continuous Control ICLR 2024
TD-MPC is a model-based reinforcement learning (RL) algorithm that performs local trajectory optimization in the latent space of a learned implicit (decoder-free) world model. In this work, we present TD-MPC2: a series of improvements upon the TD-MPC algorithm. We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perform 80 tasks across multiple task domains, embodiments, and action spaces. We conclude with an account of lessons, opportunities, and risks associated with large TD-MPC2 agents. Explore videos, models, data, code, and more at https://tdmpc2.com
comment: ICLR 2024. Explore videos, models, data, code, and more at https://tdmpc2.com
♻ ☆ TeleMoMa: A Modular and Versatile Teleoperation System for Mobile Manipulation
A critical bottleneck limiting imitation learning in robotics is the lack of data. This problem is more severe in mobile manipulation, where collecting demonstrations is harder than in stationary manipulation due to the lack of available and easy-to-use teleoperation interfaces. In this work, we demonstrate TeleMoMa, a general and modular interface for whole-body teleoperation of mobile manipulators. TeleMoMa unifies multiple human interfaces including RGB and depth cameras, virtual reality controllers, keyboard, joysticks, etc., and any combination thereof. In its more accessible version, TeleMoMa works using simply vision (e.g., an RGB-D camera), lowering the entry bar for humans to provide mobile manipulation demonstrations. We demonstrate the versatility of TeleMoMa by teleoperating several existing mobile manipulators - PAL Tiago++, Toyota HSR, and Fetch - in simulation and the real world. We demonstrate the quality of the demonstrations collected with TeleMoMa by training imitation learning policies for mobile manipulation tasks involving synchronized whole-body motion. Finally, we also show that TeleMoMa's teleoperation channel enables teleoperation on site, looking at the robot, or remote, sending commands and observations through a computer network, and perform user studies to evaluate how easy it is for novice users to learn to collect demonstrations with different combinations of human interfaces enabled by our system. We hope TeleMoMa becomes a helpful tool for the community enabling researchers to collect whole-body mobile manipulation demonstrations. For more information and video results, https://robin-lab.cs.utexas.edu/telemoma-web.
♻ ☆ OASIS: Optimal Arrangements for Sensing in SLAM
The number and arrangement of sensors on mobile robot dramatically influence its perception capabilities. Ensuring that sensors are mounted in a manner that enables accurate detection, localization, and mapping is essential for the success of downstream control tasks. However, when designing a new robotic platform, researchers and practitioners alike usually mimic standard configurations or maximize simple heuristics like field-of-view (FOV) coverage to decide where to place exteroceptive sensors. In this work, we conduct an information-theoretic investigation of this overlooked element of robotic perception in the context of simultaneous localization and mapping (SLAM). We show how to formalize the sensor arrangement problem as a form of subset selection under the E-optimality performance criterion. While this formulation is NP-hard in general, we show that a combination of greedy sensor selection and fast convex relaxation-based post-hoc verification enables the efficient recovery of certifiably optimal sensor designs in practice. Results from synthetic experiments reveal that sensors placed with OASIS outperform benchmarks in terms of mean squared error of visual SLAM estimates.
Artificial Intelligence 139
☆ MathVerse: Does Your Multi-modal LLM Truly See the Diagrams in Visual Math Problems?
The remarkable progress of Multi-modal Large Language Models (MLLMs) has garnered unparalleled attention, due to their superior performance in visual contexts. However, their capabilities in visual math problem-solving remain insufficiently evaluated and understood. We investigate current benchmarks to incorporate excessive visual content within textual questions, which potentially assist MLLMs in deducing answers without truly interpreting the input diagrams. To this end, we introduce MathVerse, an all-around visual math benchmark designed for an equitable and in-depth evaluation of MLLMs. We meticulously collect 2,612 high-quality, multi-subject math problems with diagrams from publicly available sources. Each problem is then transformed by human annotators into six distinct versions, each offering varying degrees of information content in multi-modality, contributing to 15K test samples in total. This approach allows MathVerse to comprehensively assess whether and how much MLLMs can truly understand the visual diagrams for mathematical reasoning. In addition, we propose a Chain-of-Thought (CoT) evaluation strategy for a fine-grained assessment of the output answers. Rather than naively judging True or False, we employ GPT-4(V) to adaptively extract crucial reasoning steps, and then score each step with detailed error analysis, which can reveal the intermediate CoT reasoning quality by MLLMs. We hope the MathVerse benchmark may provide unique insights to guide the future development of MLLMs. Project page: https://mathverse-cuhk.github.io
comment: 46 Pages, Work in Progress, Benchmark Project Page: https://mathverse-cuhk.github.io
☆ Videoshop: Localized Semantic Video Editing with Noise-Extrapolated Diffusion Inversion
We introduce Videoshop, a training-free video editing algorithm for localized semantic edits. Videoshop allows users to use any editing software, including Photoshop and generative inpainting, to modify the first frame; it automatically propagates those changes, with semantic, spatial, and temporally consistent motion, to the remaining frames. Unlike existing methods that enable edits only through imprecise textual instructions, Videoshop allows users to add or remove objects, semantically change objects, insert stock photos into videos, etc. with fine-grained control over locations and appearance. We achieve this through image-based video editing by inverting latents with noise extrapolation, from which we generate videos conditioned on the edited image. Videoshop produces higher quality edits against 6 baselines on 2 editing benchmarks using 10 evaluation metrics.
☆ Envisioning the Next-Generation AI Coding Assistants: Insights & Proposals
As a research-product hybrid group in AI for Software Engineering (AI4SE), we present four key takeaways from our experience developing in-IDE AI coding assistants. AI coding assistants should set clear expectations for usage, integrate with advanced IDE capabilities and existing extensions, use extendable backend designs, and collect app data responsibly for downstream analyses. We propose open questions and challenges that academia and industry should address to realize the vision of next-generation AI coding assistants.
☆ ReAct Meets ActRe: Autonomous Annotations of Agent Trajectories for Contrastive Self-Training
Language agents have demonstrated autonomous decision-making abilities by reasoning with foundation models. Recently, efforts have been made to train language agents for performance improvement, with multi-step reasoning and action trajectories as the training data. However, collecting such trajectories still requires considerable human effort, by either artificial annotations or implementations of diverse prompting frameworks. In this work, we propose A$^3$T, a framework that enables the Autonomous Annotation of Agent Trajectories in the style of ReAct. The central role is an ActRe prompting agent, which explains the reason for an arbitrary action. When randomly sampling an external action, the ReAct-style agent could query the ActRe agent with the action to obtain its textual rationales. Novel trajectories are then synthesized by prepending the posterior reasoning from ActRe to the sampled action. In this way, the ReAct-style agent executes multiple trajectories for the failed tasks, and selects the successful ones to supplement its failed trajectory for contrastive self-training. Realized by policy gradient methods with binarized rewards, the contrastive self-training with accumulated trajectories facilitates a closed loop for multiple rounds of language agent self-improvement. We conduct experiments using QLoRA fine-tuning with the open-sourced Mistral-7B-Instruct-v0.2. In AlfWorld, the agent trained with A$^3$T obtains a 1-shot success rate of 96%, and 100% success with 4 iterative rounds. In WebShop, the 1-shot performance of the A$^3$T agent matches human average, and 4 rounds of iterative refinement lead to the performance approaching human experts. A$^3$T agents significantly outperform existing techniques, including prompting with GPT-4, advanced agent frameworks, and fully fine-tuned LLMs.
☆ Large Language Models for Multi-Choice Question Classification of Medical Subjects
The aim of this paper is to evaluate whether large language models trained on multi-choice question data can be used to discriminate between medical subjects. This is an important and challenging task for automatic question answering. To achieve this goal, we train deep neural networks for multi-class classification of questions into the inferred medical subjects. Using our Multi-Question (MQ) Sequence-BERT method, we outperform the state-of-the-art results on the MedMCQA dataset with an accuracy of 0.68 and 0.60 on their development and test sets, respectively. In this sense, we show the capability of AI and LLMs in particular for multi-classification tasks in the Healthcare domain.
☆ A survey on Concept-based Approaches For Model Improvement
The focus of recent research has shifted from merely increasing the Deep Neural Networks (DNNs) performance in various tasks to DNNs, which are more interpretable to humans. The field of eXplainable Artificial Intelligence (XAI) has observed various techniques, including saliency-based and concept-based approaches. Concept-based approaches explain the model's decisions in simple human understandable terms called Concepts. Concepts are human interpretable units of data and are the thinking ground of humans. Explanations in terms of concepts enable detecting spurious correlations, inherent biases, or clever-hans. With the advent of concept-based explanations, there have been various concept representation methods and automatic concept discovery algorithms. Some recent methods use concepts for post-hoc model disentanglement evaluation, while others use them for ante-hoc training. The concept-based approaches are new, with many representations coming up, and there is very limited work on Concept-based Model improvement. We provide a systematic review and taxonomy of various concept representations and their discovery algorithms in DNNs, specifically in vision. We also provide details on concept-based model improvement literature, which is the first to survey concept-based model improvement methods.
☆ The Era of Semantic Decoding
Recent work demonstrated great promise in the idea of orchestrating collaborations between LLMs, human input, and various tools to address the inherent limitations of LLMs. We propose a novel perspective called semantic decoding, which frames these collaborative processes as optimization procedures in semantic space. Specifically, we conceptualize LLMs as semantic processors that manipulate meaningful pieces of information that we call semantic tokens (known thoughts). LLMs are among a large pool of other semantic processors, including humans and tools, such as search engines or code executors. Collectively, semantic processors engage in dynamic exchanges of semantic tokens to progressively construct high-utility outputs. We refer to these orchestrated interactions among semantic processors, optimizing and searching in semantic space, as semantic decoding algorithms. This concept draws a direct parallel to the well-studied problem of syntactic decoding, which involves crafting algorithms to best exploit auto-regressive language models for extracting high-utility sequences of syntactic tokens. By focusing on the semantic level and disregarding syntactic details, we gain a fresh perspective on the engineering of AI systems, enabling us to imagine systems with much greater complexity and capabilities. In this position paper, we formalize the transition from syntactic to semantic tokens as well as the analogy between syntactic and semantic decoding. Subsequently, we explore the possibilities of optimizing within the space of semantic tokens via semantic decoding algorithms. We conclude with a list of research opportunities and questions arising from this fresh perspective. The semantic decoding perspective offers a powerful abstraction for search and optimization directly in the space of meaningful concepts, with semantic tokens as the fundamental units of a new type of computation.
comment: 25 pages, 3 figures
☆ Lexicon-Level Contrastive Visual-Grounding Improves Language Modeling
Today's most accurate language models are trained on orders of magnitude more language data than human language learners receive - but with no supervision from other sensory modalities that play a crucial role in human learning. Can we make LMs' representations and predictions more accurate (and more human-like) with more ecologically plausible supervision? This paper describes LexiContrastive Grounding (LCG), a grounded language learning procedure that leverages visual supervision to improve textual representations. LexiContrastive Grounding combines a next token prediction strategy with a contrastive visual grounding objective, focusing on early-layer representations that encode lexical information. Across multiple word-learning and sentence-understanding benchmarks, LexiContrastive Grounding not only outperforms standard language-only models in learning efficiency, but also improves upon vision-and-language learning procedures including CLIP, GIT, Flamingo, and Vokenization. Moreover, LexiContrastive Grounding improves perplexity by around 5% on multiple language modeling tasks. This work underscores the potential of incorporating visual grounding into language models, aligning more closely with the multimodal nature of human language acquisition.
☆ Dynamic Explanation Emphasis in Human-XAI Interaction with Communication Robot
Communication robots have the potential to contribute to effective human-XAI interaction as an interface that goes beyond textual or graphical explanations. One of their strengths is that they can use physical and vocal expressions to add detailed nuances to explanations. However, it is not clear how a robot can apply such expressions, or in particular, how we can develop a strategy to adaptively use such expressions depending on the task and user in dynamic interactions. To address this question, this paper proposes DynEmph, a method for a communication robot to decide where to emphasize XAI-generated explanations with physical expressions. It predicts the effect of emphasizing certain points on a user and aims to minimize the expected difference between predicted user decisions and AI-suggested ones. DynEmph features a strategy for deciding where to emphasize in a data-driven manner, relieving engineers from the need to manually design a strategy. We further conducted experiments to investigate how emphasis selection strategies affect the performance of user decisions. The results suggest that, while a naive strategy (emphasizing explanations for an AI's most probable class) does not necessarily work better, DynEmph effectively guides users to better decisions under the condition that the performance of the AI suggestion is high.
☆ Object-Centric Domain Randomization for 3D Shape Reconstruction in the Wild
One of the biggest challenges in single-view 3D shape reconstruction in the wild is the scarcity of <3D shape, 2D image>-paired data from real-world environments. Inspired by remarkable achievements via domain randomization, we propose ObjectDR which synthesizes such paired data via a random simulation of visual variations in object appearances and backgrounds. Our data synthesis framework exploits a conditional generative model (e.g., ControlNet) to generate images conforming to spatial conditions such as 2.5D sketches, which are obtainable through a rendering process of 3D shapes from object collections (e.g., Objaverse-XL). To simulate diverse variations while preserving object silhouettes embedded in spatial conditions, we also introduce a disentangled framework which leverages an initial object guidance. After synthesizing a wide range of data, we pre-train a model on them so that it learns to capture a domain-invariant geometry prior which is consistent across various domains. We validate its effectiveness by substantially improving 3D shape reconstruction models on a real-world benchmark. In a scale-up evaluation, our pre-training achieves 23.6% superior results compared with the pre-training on high-quality computer graphics renderings.
comment: Project Page: https://ObjectDR.github.io
☆ Click to Grasp: Zero-Shot Precise Manipulation via Visual Diffusion Descriptors
Precise manipulation that is generalizable across scenes and objects remains a persistent challenge in robotics. Current approaches for this task heavily depend on having a significant number of training instances to handle objects with pronounced visual and/or geometric part ambiguities. Our work explores the grounding of fine-grained part descriptors for precise manipulation in a zero-shot setting by utilizing web-trained text-to-image diffusion-based generative models. We tackle the problem by framing it as a dense semantic part correspondence task. Our model returns a gripper pose for manipulating a specific part, using as reference a user-defined click from a source image of a visually different instance of the same object. We require no manual grasping demonstrations as we leverage the intrinsic object geometry and features. Practical experiments in a real-world tabletop scenario validate the efficacy of our approach, demonstrating its potential for advancing semantic-aware robotics manipulation. Web page: https://tsagkas.github.io/click2grasp
comment: 8 pages, 4 figures
☆ Constrained Reinforcement Learning with Smoothed Log Barrier Function
Reinforcement Learning (RL) has been widely applied to many control tasks and substantially improved the performances compared to conventional control methods in many domains where the reward function is well defined. However, for many real-world problems, it is often more convenient to formulate optimization problems in terms of rewards and constraints simultaneously. Optimizing such constrained problems via reward shaping can be difficult as it requires tedious manual tuning of reward functions with several interacting terms. Recent formulations which include constraints mostly require a pre-training phase, which often needs human expertise to collect data or assumes having a sub-optimal policy readily available. We propose a new constrained RL method called CSAC-LB (Constrained Soft Actor-Critic with Log Barrier Function), which achieves competitive performance without any pre-training by applying a linear smoothed log barrier function to an additional safety critic. It implements an adaptive penalty for policy learning and alleviates the numerical issues that are known to complicate the application of the log barrier function method. As a result, we show that with CSAC-LB, we achieve state-of-the-art performance on several constrained control tasks with different levels of difficulty and evaluate our methods in a locomotion task on a real quadruped robot platform.
☆ Soft Learning Probabilistic Circuits
Probabilistic Circuits (PCs) are prominent tractable probabilistic models, allowing for a range of exact inferences. This paper focuses on the main algorithm for training PCs, LearnSPN, a gold standard due to its efficiency, performance, and ease of use, in particular for tabular data. We show that LearnSPN is a greedy likelihood maximizer under mild assumptions. While inferences in PCs may use the entire circuit structure for processing queries, LearnSPN applies a hard method for learning them, propagating at each sum node a data point through one and only one of the children/edges as in a hard clustering process. We propose a new learning procedure named SoftLearn, that induces a PC using a soft clustering process. We investigate the effect of this learning-inference compatibility in PCs. Our experiments show that SoftLearn outperforms LearnSPN in many situations, yielding better likelihoods and arguably better samples. We also analyze comparable tractable models to highlight the differences between soft/hard learning and model querying.
☆ How Human-Centered Explainable AI Interface Are Designed and Evaluated: A Systematic Survey
Despite its technological breakthroughs, eXplainable Artificial Intelligence (XAI) research has limited success in producing the {\em effective explanations} needed by users. In order to improve XAI systems' usability, practical interpretability, and efficacy for real users, the emerging area of {\em Explainable Interfaces} (EIs) focuses on the user interface and user experience design aspects of XAI. This paper presents a systematic survey of 53 publications to identify current trends in human-XAI interaction and promising directions for EI design and development. This is among the first systematic survey of EI research.
☆ Learning to Project for Cross-Task Knowledge Distillation
Traditional knowledge distillation (KD) relies on a proficient teacher trained on the target task, which is not always available. In this setting, cross-task distillation can be used, enabling the use of any teacher model trained on a different task. However, many KD methods prove ineffective when applied to this cross-task setting. To address this limitation, we propose a simple modification: the use of an inverted projection. We show that this drop-in replacement for a standard projector is effective by learning to disregard any task-specific features which might degrade the student's performance. We find that this simple modification is sufficient for extending many KD methods to the cross-task setting, where the teacher and student tasks can be very different. In doing so, we obtain up to a 1.9% improvement in the cross-task setting compared to the traditional projection, at no additional cost. Our method can obtain significant performance improvements (up to 7%) when using even a randomly-initialised teacher on various tasks such as depth estimation, image translation, and semantic segmentation, despite the lack of any learned knowledge to transfer. To provide conceptual and analytical insights into this result, we show that using an inverted projection allows the distillation loss to be decomposed into a knowledge transfer and a spectral regularisation component. Through this analysis we are additionally able to propose a novel regularisation loss that allows teacher-free distillation, enabling performance improvements of up to 8.57% on ImageNet with no additional training costs.
☆ Physics-Based Causal Reasoning for Safe & Robust Next-Best Action Selection in Robot Manipulation Tasks IROS
Safe and efficient object manipulation is a key enabler of many real-world robot applications. However, this is challenging because robot operation must be robust to a range of sensor and actuator uncertainties. In this paper, we present a physics-informed causal-inference-based framework for a robot to probabilistically reason about candidate actions in a block stacking task in a partially observable setting. We integrate a physics-based simulation of the rigid-body system dynamics with a causal Bayesian network (CBN) formulation to define a causal generative probabilistic model of the robot decision-making process. Using simulation-based Monte Carlo experiments, we demonstrate our framework's ability to successfully: (1) predict block tower stability with high accuracy (Pred Acc: 88.6%); and, (2) select an approximate next-best action for the block stacking task, for execution by an integrated robot system, achieving 94.2% task success rate. We also demonstrate our framework's suitability for real-world robot systems by demonstrating successful task executions with a domestic support robot, with perception and manipulation sub-system integration. Hence, we show that by embedding physics-based causal reasoning into robots' decision-making processes, we can make robot task execution safer, more reliable, and more robust to various types of uncertainty.
comment: 8 pages, 9 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
☆ HyperGALE: ASD Classification via Hypergraph Gated Attention with Learnable Hyperedges IJCNN 2024
Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by varied social cognitive challenges and repetitive behavioral patterns. Identifying reliable brain imaging-based biomarkers for ASD has been a persistent challenge due to the spectrum's diverse symptomatology. Existing baselines in the field have made significant strides in this direction, yet there remains room for improvement in both performance and interpretability. We propose \emph{HyperGALE}, which builds upon the hypergraph by incorporating learned hyperedges and gated attention mechanisms. This approach has led to substantial improvements in the model's ability to interpret complex brain graph data, offering deeper insights into ASD biomarker characterization. Evaluated on the extensive ABIDE II dataset, \emph{HyperGALE} not only improves interpretability but also demonstrates statistically significant enhancements in key performance metrics compared to both previous baselines and the foundational hypergraph model. The advancement \emph{HyperGALE} brings to ASD research highlights the potential of sophisticated graph-based techniques in neurodevelopmental studies. The source code and implementation instructions are available at GitHub:https://github.com/mehular0ra/HyperGALE.
comment: Accepted to IJCNN 2024
☆ Utilizing the LightGBM Algorithm for Operator User Credit Assessment Research
Mobile Internet user credit assessment is an important way for communication operators to establish decisions and formulate measures, and it is also a guarantee for operators to obtain expected benefits. However, credit evaluation methods have long been monopolized by financial industries such as banks and credit. As supporters and providers of platform network technology and network resources, communication operators are also builders and maintainers of communication networks. Internet data improves the user's credit evaluation strategy. This paper uses the massive data provided by communication operators to carry out research on the operator's user credit evaluation model based on the fusion LightGBM algorithm. First, for the massive data related to user evaluation provided by operators, key features are extracted by data preprocessing and feature engineering methods, and a multi-dimensional feature set with statistical significance is constructed; then, linear regression, decision tree, LightGBM, and other machine learning algorithms build multiple basic models to find the best basic model; finally, integrates Averaging, Voting, Blending, Stacking and other integrated algorithms to refine multiple fusion models, and finally establish the most suitable fusion model for operator user evaluation.
☆ Detoxifying Large Language Models via Knowledge Editing
This paper investigates using knowledge editing techniques to detoxify Large Language Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories with various powerful attack prompts and equips comprehensive metrics for systematic evaluation. We conduct experiments to compare knowledge editing approaches with previous baselines, indicating that knowledge editing has the potential to efficiently detoxify LLMs with limited impact on general performance. Then, we propose a simple yet effective baseline, dubbed Detoxifying with Intraoperative Neural Monitoring (DINM), to diminish the toxicity of LLMs within a few tuning steps via only one instance. We further provide an in-depth analysis of the internal mechanism for various detoxify approaches, demonstrating that previous methods like SFT and DPO may merely suppress the activations of toxic parameters, while DINM mitigates the toxicity of the toxic parameters to a certain extent, making permanent adjustments. We hope that these insights could shed light on future work of developing detoxifying approaches and the underlying knowledge mechanisms of LLMs. Code and benchmark are available at https://github.com/zjunlp/EasyEdit.
comment: Ongoing work. Project website: https://zjunlp.github.io/project/SafeEdit Benchmark: https://huggingface.co/datasets/zjunlp/SafeEdit Code: https://github.com/zjunlp/EasyEdit
☆ ChatGPT Alternative Solutions: Large Language Models Survey
In recent times, the grandeur of Large Language Models (LLMs) has not only shone in the realm of natural language processing but has also cast its brilliance across a vast array of applications. This remarkable display of LLM capabilities has ignited a surge in research contributions within this domain, spanning a diverse spectrum of topics. These contributions encompass advancements in neural network architecture, context length enhancements, model alignment, training datasets, benchmarking, efficiency improvements, and more. Recent years have witnessed a dynamic synergy between academia and industry, propelling the field of LLM research to new heights. A notable milestone in this journey is the introduction of ChatGPT, a powerful AI chatbot grounded in LLMs, which has garnered widespread societal attention. The evolving technology of LLMs has begun to reshape the landscape of the entire AI community, promising a revolutionary shift in the way we create and employ AI algorithms. Given this swift-paced technical evolution, our survey embarks on a journey to encapsulate the recent strides made in the world of LLMs. Through an exploration of the background, key discoveries, and prevailing methodologies, we offer an up-to-the-minute review of the literature. By examining multiple LLM models, our paper not only presents a comprehensive overview but also charts a course that identifies existing challenges and points toward potential future research trajectories. This survey furnishes a well-rounded perspective on the current state of generative AI, shedding light on opportunities for further exploration, enhancement, and innovation.
☆ AnyV2V: A Plug-and-Play Framework For Any Video-to-Video Editing Tasks
Video-to-video editing involves editing a source video along with additional control (such as text prompts, subjects, or styles) to generate a new video that aligns with the source video and the provided control. Traditional methods have been constrained to certain editing types, limiting their ability to meet the wide range of user demands. In this paper, we introduce AnyV2V, a novel training-free framework designed to simplify video editing into two primary steps: (1) employing an off-the-shelf image editing model (e.g. InstructPix2Pix, InstantID, etc) to modify the first frame, (2) utilizing an existing image-to-video generation model (e.g. I2VGen-XL) for DDIM inversion and feature injection. In the first stage, AnyV2V can plug in any existing image editing tools to support an extensive array of video editing tasks. Beyond the traditional prompt-based editing methods, AnyV2V also can support novel video editing tasks, including reference-based style transfer, subject-driven editing, and identity manipulation, which were unattainable by previous methods. In the second stage, AnyV2V can plug in any existing image-to-video models to perform DDIM inversion and intermediate feature injection to maintain the appearance and motion consistency with the source video. On the prompt-based editing, we show that AnyV2V can outperform the previous best approach by 35\% on prompt alignment, and 25\% on human preference. On the three novel tasks, we show that AnyV2V also achieves a high success rate. We believe AnyV2V will continue to thrive due to its ability to seamlessly integrate the fast-evolving image editing methods. Such compatibility can help AnyV2V to increase its versatility to cater to diverse user demands.
comment: preprint
☆ Towards Single-System Illusion in Software-Defined Vehicles -- Automated, AI-Powered Workflow
We propose a novel model- and feature-based approach to development of vehicle software systems, where the end architecture is not explicitly defined. Instead, it emerges from an iterative process of search and optimization given certain constraints, requirements and hardware architecture, while retaining the property of single-system illusion, where applications run in a logically uniform environment. One of the key points of the presented approach is the inclusion of modern generative AI, specifically Large Language Models (LLMs), in the loop. With the recent advances in the field, we expect that the LLMs will be able to assist in processing of requirements, generation of formal system models, as well as generation of software deployment specification and test code. The resulting pipeline is automated to a large extent, with feedback being generated at each step.
☆ Multi-Level Explanations for Generative Language Models
Perturbation-based explanation methods such as LIME and SHAP are commonly applied to text classification. This work focuses on their extension to generative language models. To address the challenges of text as output and long text inputs, we propose a general framework called MExGen that can be instantiated with different attribution algorithms. To handle text output, we introduce the notion of scalarizers for mapping text to real numbers and investigate multiple possibilities. To handle long inputs, we take a multi-level approach, proceeding from coarser levels of granularity to finer ones, and focus on algorithms with linear scaling in model queries. We conduct a systematic evaluation, both automated and human, of perturbation-based attribution methods for summarization and context-grounded question answering. The results show that our framework can provide more locally faithful explanations of generated outputs.
☆ Language Models Can Reduce Asymmetry in Information Markets
This work addresses the buyer's inspection paradox for information markets. The paradox is that buyers need to access information to determine its value, while sellers need to limit access to prevent theft. To study this, we introduce an open-source simulated digital marketplace where intelligent agents, powered by language models, buy and sell information on behalf of external participants. The central mechanism enabling this marketplace is the agents' dual capabilities: they not only have the capacity to assess the quality of privileged information but also come equipped with the ability to forget. This ability to induce amnesia allows vendors to grant temporary access to proprietary information, significantly reducing the risk of unauthorized retention while enabling agents to accurately gauge the information's relevance to specific queries or tasks. To perform well, agents must make rational decisions, strategically explore the marketplace through generated sub-queries, and synthesize answers from purchased information. Concretely, our experiments (a) uncover biases in language models leading to irrational behavior and evaluate techniques to mitigate these biases, (b) investigate how price affects demand in the context of informational goods, and (c) show that inspection and higher budgets both lead to higher quality outcomes.
☆ Analysing Diffusion Segmentation for Medical Images
Denoising Diffusion Probabilistic models have become increasingly popular due to their ability to offer probabilistic modeling and generate diverse outputs. This versatility inspired their adaptation for image segmentation, where multiple predictions of the model can produce segmentation results that not only achieve high quality but also capture the uncertainty inherent in the model. Here, powerful architectures were proposed for improving diffusion segmentation performance. However, there is a notable lack of analysis and discussions on the differences between diffusion segmentation and image generation, and thorough evaluations are missing that distinguish the improvements these architectures provide for segmentation in general from their benefit for diffusion segmentation specifically. In this work, we critically analyse and discuss how diffusion segmentation for medical images differs from diffusion image generation, with a particular focus on the training behavior. Furthermore, we conduct an assessment how proposed diffusion segmentation architectures perform when trained directly for segmentation. Lastly, we explore how different medical segmentation tasks influence the diffusion segmentation behavior and the diffusion process could be adapted accordingly. With these analyses, we aim to provide in-depth insights into the behavior of diffusion segmentation that allow for a better design and evaluation of diffusion segmentation methods in the future.
☆ Biased Binary Attribute Classifiers Ignore the Majority Classes
To visualize the regions of interest that classifiers base their decisions on, different Class Activation Mapping (CAM) methods have been developed. However, all of these techniques target categorical classifiers only, though most real-world tasks are binary classification. In this paper, we extend gradient-based CAM techniques to work with binary classifiers and visualize the active regions for binary facial attribute classifiers. When training an unbalanced binary classifier on an imbalanced dataset, it is well-known that the majority class, i.e. the class with many training samples, is mostly predicted much better than minority class with few training instances. In our experiments on the CelebA dataset, we verify these results, when training an unbalanced classifier to extract 40 facial attributes simultaneously. One would expect that the biased classifier has learned to extract features mainly for the majority classes and that the proportional energy of the activations mainly reside in certain specific regions of the image where the attribute is located. However, we find very little regular activation for samples of majority classes, while the active regions for minority classes seem mostly reasonable and overlap with our expectations. These results suggest that biased classifiers mainly rely on bias activation for majority classes. When training a balanced classifier on the imbalanced data by employing attribute-specific class weights, majority and minority classes are classified similarly well and show expected activations for almost all attributes
☆ On the continuity and smoothness of the value function in reinforcement learning and optimal control
The value function plays a crucial role as a measure for the cumulative future reward an agent receives in both reinforcement learning and optimal control. It is therefore of interest to study how similar the values of neighboring states are, i.e., to investigate the continuity of the value function. We do so by providing and verifying upper bounds on the value function's modulus of continuity. Additionally, we show that the value function is always H\"older continuous under relatively weak assumptions on the underlying system and that non-differentiable value functions can be made differentiable by slightly "disturbing" the system.
☆ Style-Extracting Diffusion Models for Semi-Supervised Histopathology Segmentation
Deep learning-based image generation has seen significant advancements with diffusion models, notably improving the quality of generated images. Despite these developments, generating images with unseen characteristics beneficial for downstream tasks has received limited attention. To bridge this gap, we propose Style-Extracting Diffusion Models, featuring two conditioning mechanisms. Specifically, we utilize 1) a style conditioning mechanism which allows to inject style information of previously unseen images during image generation and 2) a content conditioning which can be targeted to a downstream task, e.g., layout for segmentation. We introduce a trainable style encoder to extract style information from images, and an aggregation block that merges style information from multiple style inputs. This architecture enables the generation of images with unseen styles in a zero-shot manner, by leveraging styles from unseen images, resulting in more diverse generations. In this work, we use the image layout as target condition and first show the capability of our method on a natural image dataset as a proof-of-concept. We further demonstrate its versatility in histopathology, where we combine prior knowledge about tissue composition and unannotated data to create diverse synthetic images with known layouts. This allows us to generate additional synthetic data to train a segmentation network in a semi-supervised fashion. We verify the added value of the generated images by showing improved segmentation results and lower performance variability between patients when synthetic images are included during segmentation training. Our code will be made publicly available at [LINK].
☆ GLC++: Source-Free Universal Domain Adaptation through Global-Local Clustering and Contrastive Affinity Learning CVPR 2023
Deep neural networks often exhibit sub-optimal performance under covariate and category shifts. Source-Free Domain Adaptation (SFDA) presents a promising solution to this dilemma, yet most SFDA approaches are restricted to closed-set scenarios. In this paper, we explore Source-Free Universal Domain Adaptation (SF-UniDA) aiming to accurately classify "known" data belonging to common categories and segregate them from target-private "unknown" data. We propose a novel Global and Local Clustering (GLC) technique, which comprises an adaptive one-vs-all global clustering algorithm to discern between target classes, complemented by a local k-NN clustering strategy to mitigate negative transfer. Despite the effectiveness, the inherent closed-set source architecture leads to uniform treatment of "unknown" data, impeding the identification of distinct "unknown" categories. To address this, we evolve GLC to GLC++, integrating a contrastive affinity learning strategy. We examine the superiority of GLC and GLC++ across multiple benchmarks and category shift scenarios. Remarkably, in the most challenging open-partial-set scenarios, GLC and GLC++ surpass GATE by 16.7% and 18.6% in H-score on VisDA, respectively. GLC++ enhances the novel category clustering accuracy of GLC by 4.3% in open-set scenarios on Office-Home. Furthermore, the introduced contrastive learning strategy not only enhances GLC but also significantly facilitates existing methodologies.
comment: This is a substantial extension of the CVPR 2023 paper "Upcycling Models under Domain and Category Shift"
☆ Locating and Mitigating Gender Bias in Large Language Models
Large language models(LLM) are pre-trained on extensive corpora to learn facts and human cognition which contain human preferences. However, this process can inadvertently lead to these models acquiring biases and stereotypes prevalent in society. Prior research has typically tackled the issue of bias through a one-dimensional perspective, concentrating either on locating or mitigating it. This limited perspective has created obstacles in facilitating research on bias to synergistically complement and progressively build upon one another. In this study, we integrate the processes of locating and mitigating bias within a unified framework. Initially, we use causal mediation analysis to trace the causal effects of different components' activation within a large language model. Building on this, we propose the LSDM (Least Square Debias Method), a knowledge-editing based method for mitigating gender bias in occupational pronouns, and compare it against two baselines on three gender bias datasets and seven knowledge competency test datasets. The experimental results indicate that the primary contributors to gender bias are the bottom MLP modules acting on the last token of occupational pronouns and the top attention module acting on the final word in the sentence. Furthermore, LSDM mitigates gender bias in the model more effectively than the other baselines, while fully preserving the model's capabilities in all other aspects.
comment: 23 pages, 5 figures
☆ Adaptive-RAG: Learning to Adapt Retrieval-Augmented Large Language Models through Question Complexity NAACL 2024
Retrieval-Augmented Large Language Models (LLMs), which incorporate the non-parametric knowledge from external knowledge bases into LLMs, have emerged as a promising approach to enhancing response accuracy in several tasks, such as Question-Answering (QA). However, even though there are various approaches dealing with queries of different complexities, they either handle simple queries with unnecessary computational overhead or fail to adequately address complex multi-step queries; yet, not all user requests fall into only one of the simple or complex categories. In this work, we propose a novel adaptive QA framework, that can dynamically select the most suitable strategy for (retrieval-augmented) LLMs from the simplest to the most sophisticated ones based on the query complexity. Also, this selection process is operationalized with a classifier, which is a smaller LM trained to predict the complexity level of incoming queries with automatically collected labels, obtained from actual predicted outcomes of models and inherent inductive biases in datasets. This approach offers a balanced strategy, seamlessly adapting between the iterative and single-step retrieval-augmented LLMs, as well as the no-retrieval methods, in response to a range of query complexities. We validate our model on a set of open-domain QA datasets, covering multiple query complexities, and show that ours enhances the overall efficiency and accuracy of QA systems, compared to relevant baselines including the adaptive retrieval approaches. Code is available at: https://github.com/starsuzi/Adaptive-RAG.
comment: NAACL 2024
☆ Building Accurate Translation-Tailored LLMs with Language Aware Instruction Tuning
Translation-tailored Large language models (LLMs) exhibit remarkable translation capabilities, even competing with supervised-trained commercial translation systems. However, off-target translation remains an unsolved problem, especially for low-resource languages, hindering us from developing accurate LLMs-based translation models. To mitigate the off-target translation problem and enhance the performance of LLMs on translation, recent works have either designed advanced prompting strategies to highlight the functionality of translation instructions or exploited the in-context learning ability of LLMs by feeding few-shot demonstrations. However, these methods essentially do not improve LLM's ability to follow translation instructions, especially the language direction information. In this work, we design a two-stage fine-tuning algorithm to improve the instruction-following ability (especially the translation direction) of LLMs. Specifically, we first tune LLMs with the maximum likelihood estimation loss on the translation dataset to elicit the basic translation capabilities. In the second stage, we construct instruction-conflicting samples by randomly replacing the translation directions with a wrong one within the instruction, and then introduce an extra unlikelihood loss to learn those samples. Experiments on IWSLT and WMT benchmarks upon the LLaMA model spanning 16 zero-shot directions show that, compared to the competitive baseline -- translation-finetuned LLama, our method could effectively reduce the off-target translation ratio (averagely -53.3\%), thus improving translation quality with average +5.7 SacreBLEU and +16.4 BLEURT. Analysis shows that our method could preserve the model's general task performance on AlpacaEval. Code and models will be released at \url{https://github.com/alphadl/LanguageAware_Tuning}.
☆ Editing Knowledge Representation of Language Lodel via Rephrased Prefix Prompts
Neural language models (LMs) have been extensively trained on vast corpora to store factual knowledge about various aspects of the world described in texts. Current technologies typically employ knowledge editing methods or specific prompts to modify LM outputs. However, existing knowledge editing methods are costly and inefficient, struggling to produce appropriate text. Additionally, prompt engineering is opaque and requires significant effort to find suitable prompts. To address these issues, we introduce a new method called PSPEM (Prefix Soft Prompt Editing Method), that can be used for a lifetime with just one training. It resolves the inefficiencies and generalizability issues in knowledge editing methods and overcomes the opacity of prompt engineering by automatically seeking optimal soft prompts. Specifically, PSPEM utilizes a prompt encoder and an encoding converter to refine key information in prompts and uses prompt alignment techniques to guide model generation, ensuring text consistency and adherence to the intended structure and content, thereby maintaining an optimal balance between efficiency and accuracy. We have validated the effectiveness of PSPEM through knowledge editing and attribute inserting. On the COUNTERFACT dataset, PSPEM achieved nearly 100\% editing accuracy and demonstrated the highest level of fluency. We further analyzed the similarities between PSPEM and original prompts and their impact on the model's internals. The results indicate that PSPEM can serve as an alternative to original prompts, supporting the model in effective editing.
comment: 19pages,3figures
☆ Loop Improvement: An Efficient Approach for Extracting Shared Features from Heterogeneous Data without Central Server
In federated learning, data heterogeneity significantly impacts performance. A typical solution involves segregating these parameters into shared and personalized components, a concept also relevant in multi-task learning. Addressing this, we propose "Loop Improvement" (LI), a novel method enhancing this separation and feature extraction without necessitating a central server or data interchange among participants. Our experiments reveal LI's superiority in several aspects: In personalized federated learning environments, LI consistently outperforms the advanced FedALA algorithm in accuracy across diverse scenarios. Additionally, LI's feature extractor closely matches the performance achieved when aggregating data from all clients. In global model contexts, employing LI with stacked personalized layers and an additional network also yields comparable results to combined client data scenarios. Furthermore, LI's adaptability extends to multi-task learning, streamlining the extraction of common features across tasks and obviating the need for simultaneous training. This approach not only enhances individual task performance but also achieves accuracy levels on par with classic multi-task learning methods where all tasks are trained simultaneously. LI integrates a loop topology with layer-wise and end-to-end training, compatible with various neural network models. This paper also delves into the theoretical underpinnings of LI's effectiveness, offering insights into its potential applications. The code is on https://github.com/axedge1983/LI
comment: 11 pages, 11 figures
Exploring the Potential of Large Language Models in Graph Generation
Large language models (LLMs) have achieved great success in many fields, and recent works have studied exploring LLMs for graph discriminative tasks such as node classification. However, the abilities of LLMs for graph generation remain unexplored in the literature. Graph generation requires the LLM to generate graphs with given properties, which has valuable real-world applications such as drug discovery, while tends to be more challenging. In this paper, we propose LLM4GraphGen to explore the ability of LLMs for graph generation with systematical task designs and extensive experiments. Specifically, we propose several tasks tailored with comprehensive experiments to address key questions regarding LLMs' understanding of different graph structure rules, their ability to capture structural type distributions, and their utilization of domain knowledge for property-based graph generation. Our evaluations demonstrate that LLMs, particularly GPT-4, exhibit preliminary abilities in graph generation tasks, including rule-based and distribution-based generation. We also observe that popular prompting methods, such as few-shot and chain-of-thought prompting, do not consistently enhance performance. Besides, LLMs show potential in generating molecules with specific properties. These findings may serve as foundations for designing good LLMs based models for graph generation and provide valuable insights and further research.
Exploring Task Unification in Graph Representation Learning via Generative Approach
Graphs are ubiquitous in real-world scenarios and encompass a diverse range of tasks, from node-, edge-, and graph-level tasks to transfer learning. However, designing specific tasks for each type of graph data is often costly and lacks generalizability. Recent endeavors under the "Pre-training + Fine-tuning" or "Pre-training + Prompt" paradigms aim to design a unified framework capable of generalizing across multiple graph tasks. Among these, graph autoencoders (GAEs), generative self-supervised models, have demonstrated their potential in effectively addressing various graph tasks. Nevertheless, these methods typically employ multi-stage training and require adaptive designs, which on one hand make it difficult to be seamlessly applied to diverse graph tasks and on the other hand overlook the negative impact caused by discrepancies in task objectives between the different stages. To address these challenges, we propose GA^2E, a unified adversarially masked autoencoder capable of addressing the above challenges seamlessly. Specifically, GA^2E proposes to use the subgraph as the meta-structure, which remains consistent across all graph tasks (ranging from node-, edge-, and graph-level to transfer learning) and all stages (both during training and inference). Further, GA^2E operates in a \textbf{"Generate then Discriminate"} manner. It leverages the masked GAE to reconstruct the input subgraph whilst treating it as a generator to compel the reconstructed graphs resemble the input subgraph. Furthermore, GA^2E introduces an auxiliary discriminator to discern the authenticity between the reconstructed (generated) subgraph and the input subgraph, thus ensuring the robustness of the graph representation through adversarial training mechanisms. We validate GA^2E's capabilities through extensive experiments on 21 datasets across four types of graph tasks.
☆ $\nabla τ$: Gradient-based and Task-Agnostic machine Unlearning
Machine Unlearning, the process of selectively eliminating the influence of certain data examples used during a model's training, has gained significant attention as a means for practitioners to comply with recent data protection regulations. However, existing unlearning methods face critical drawbacks, including their prohibitively high cost, often associated with a large number of hyperparameters, and the limitation of forgetting only relatively small data portions. This often makes retraining the model from scratch a quicker and more effective solution. In this study, we introduce Gradient-based and Task-Agnostic machine Unlearning ($\nabla \tau$), an optimization framework designed to remove the influence of a subset of training data efficiently. It applies adaptive gradient ascent to the data to be forgotten while using standard gradient descent for the remaining data. $\nabla \tau$ offers multiple benefits over existing approaches. It enables the unlearning of large sections of the training dataset (up to 30%). It is versatile, supporting various unlearning tasks (such as subset forgetting or class removal) and applicable across different domains (images, text, etc.). Importantly, $\nabla \tau$ requires no hyperparameter adjustments, making it a more appealing option than retraining the model from scratch. We evaluate our framework's effectiveness using a set of well-established Membership Inference Attack metrics, demonstrating up to 10% enhancements in performance compared to state-of-the-art methods without compromising the original model's accuracy.
comment: 14 pages, 2 figures
☆ Distilling Reinforcement Learning Policies for Interpretable Robot Locomotion: Gradient Boosting Machines and Symbolic Regression
Recent advancements in reinforcement learning (RL) have led to remarkable achievements in robot locomotion capabilities. However, the complexity and ``black-box'' nature of neural network-based RL policies hinder their interpretability and broader acceptance, particularly in applications demanding high levels of safety and reliability. This paper introduces a novel approach to distill neural RL policies into more interpretable forms using Gradient Boosting Machines (GBMs), Explainable Boosting Machines (EBMs) and Symbolic Regression. By leveraging the inherent interpretability of generalized additive models, decision trees, and analytical expressions, we transform opaque neural network policies into more transparent ``glass-box'' models. We train expert neural network policies using RL and subsequently distill them into (i) GBMs, (ii) EBMs, and (iii) symbolic policies. To address the inherent distribution shift challenge of behavioral cloning, we propose to use the Dataset Aggregation (DAgger) algorithm with a curriculum of episode-dependent alternation of actions between expert and distilled policies, to enable efficient distillation of feedback control policies. We evaluate our approach on various robot locomotion gaits -- walking, trotting, bounding, and pacing -- and study the importance of different observations in joint actions for distilled policies using various methods. We train neural expert policies for 205 hours of simulated experience and distill interpretable policies with only 10 minutes of simulated interaction for each gait using the proposed method.
☆ DexDribbler: Learning Dexterous Soccer Manipulation via Dynamic Supervision IROS 2024
Learning dexterous locomotion policy for legged robots is becoming increasingly popular due to its ability to handle diverse terrains and resemble intelligent behaviors. However, joint manipulation of moving objects and locomotion with legs, such as playing soccer, receive scant attention in the learning community, although it is natural for humans and smart animals. A key challenge to solve this multitask problem is to infer the objectives of locomotion from the states and targets of the manipulated objects. The implicit relation between the object states and robot locomotion can be hard to capture directly from the training experience. We propose adding a feedback control block to compute the necessary body-level movement accurately and using the outputs as dynamic joint-level locomotion supervision explicitly. We further utilize an improved ball dynamic model, an extended context-aided estimator, and a comprehensive ball observer to facilitate transferring policy learned in simulation to the real world. We observe that our learning scheme can not only make the policy network converge faster but also enable soccer robots to perform sophisticated maneuvers like sharp cuts and turns on flat surfaces, a capability that was lacking in previous methods. Video and code are available at https://github.com/SysCV/soccer-player
comment: 8 pages, 7 figures, submitted to IROS 2024
☆ From Perils to Possibilities: Understanding how Human (and AI) Biases affect Online Fora
Social media platforms are online fora where users engage in discussions, share content, and build connections. This review explores the dynamics of social interactions, user-generated contents, and biases within the context of social media analysis (analyzing works that use the tools offered by complex network analysis and natural language processing) through the lens of three key points of view: online debates, online support, and human-AI interactions. On the one hand, we delineate the phenomenon of online debates, where polarization, misinformation, and echo chamber formation often proliferate, driven by algorithmic biases and extreme mechanisms of homophily. On the other hand, we explore the emergence of online support groups through users' self-disclosure and social support mechanisms. Online debates and support mechanisms present a duality of both perils and possibilities within social media; perils of segregated communities and polarized debates, and possibilities of empathy narratives and self-help groups. This dichotomy also extends to a third perspective: users' reliance on AI-generated content, such as the ones produced by Large Language Models, which can manifest both human biases hidden in training sets and non-human biases that emerge from their artificial neural architectures. Analyzing interdisciplinary approaches, we aim to deepen the understanding of the complex interplay between social interactions, user-generated content, and biases within the realm of social media ecosystems.
☆ Impact Assessment of Missing Data in Model Predictions for Earth Observation Applications
Earth observation (EO) applications involving complex and heterogeneous data sources are commonly approached with machine learning models. However, there is a common assumption that data sources will be persistently available. Different situations could affect the availability of EO sources, like noise, clouds, or satellite mission failures. In this work, we assess the impact of missing temporal and static EO sources in trained models across four datasets with classification and regression tasks. We compare the predictive quality of different methods and find that some are naturally more robust to missing data. The Ensemble strategy, in particular, achieves a prediction robustness up to 100%. We evidence that missing scenarios are significantly more challenging in regression than classification tasks. Finally, we find that the optical view is the most critical view when it is missing individually.
comment: Accepted at IEEE International Geoscience and Remote Sensing Symposium 2024
☆ Enhancing Historical Image Retrieval with Compositional Cues
In analyzing vast amounts of digitally stored historical image data, existing content-based retrieval methods often overlook significant non-semantic information, limiting their effectiveness for flexible exploration across varied themes. To broaden the applicability of image retrieval methods for diverse purposes and uncover more general patterns, we innovatively introduce a crucial factor from computational aesthetics, namely image composition, into this topic. By explicitly integrating composition-related information extracted by CNN into the designed retrieval model, our method considers both the image's composition rules and semantic information. Qualitative and quantitative experiments demonstrate that the image retrieval network guided by composition information outperforms those relying solely on content information, facilitating the identification of images in databases closer to the target image in human perception. Please visit https://github.com/linty5/CCBIR to try our codes.
☆ How to be fair? A study of label and selection bias
It is widely accepted that biased data leads to biased and thus potentially unfair models. Therefore, several measures for bias in data and model predictions have been proposed, as well as bias mitigation techniques whose aim is to learn models that are fair by design. Despite the myriad of mitigation techniques developed in the past decade, however, it is still poorly understood under what circumstances which methods work. Recently, Wick et al. showed, with experiments on synthetic data, that there exist situations in which bias mitigation techniques lead to more accurate models when measured on unbiased data. Nevertheless, in the absence of a thorough mathematical analysis, it remains unclear which techniques are effective under what circumstances. We propose to address this problem by establishing relationships between the type of bias and the effectiveness of a mitigation technique, where we categorize the mitigation techniques by the bias measure they optimize. In this paper we illustrate this principle for label and selection bias on the one hand, and demographic parity and ``We're All Equal'' on the other hand. Our theoretical analysis allows to explain the results of Wick et al. and we also show that there are situations where minimizing fairness measures does not result in the fairest possible distribution.
☆ Multi-role Consensus through LLMs Discussions for Vulnerability Detection
Recent advancements in large language models (LLMs) have highlighted the potential for vulnerability detection, a crucial component of software quality assurance. Despite this progress, most studies have been limited to the perspective of a single role, usually testers, lacking diverse viewpoints from different roles in a typical software development life-cycle, including both developers and testers. To this end, this paper introduces an approach to employ LLMs to act as different roles to simulate real-life code review process, engaging in discussions towards a consensus on the existence and classification of vulnerabilities in the code. Preliminary evaluation of the proposed approach indicates a 4.73% increase in the precision rate, 58.9% increase in the recall rate, and a 28.1% increase in the F1 score.
☆ Reactor Optimization Benchmark by Reinforcement Learning
Neutronic calculations for reactors are a daunting task when using Monte Carlo (MC) methods. As high-performance computing has advanced, the simulation of a reactor is nowadays more readily done, but design and optimization with multiple parameters is still a computational challenge. MC transport simulations, coupled with machine learning techniques, offer promising avenues for enhancing the efficiency and effectiveness of nuclear reactor optimization. This paper introduces a novel benchmark problem within the OpenNeoMC framework designed specifically for reinforcement learning. The benchmark involves optimizing a unit cell of a research reactor with two varying parameters (fuel density and water spacing) to maximize neutron flux while maintaining reactor criticality. The test case features distinct local optima, representing different physical regimes, thus posing a challenge for learning algorithms. Through extensive simulations utilizing evolutionary and neuroevolutionary algorithms, we demonstrate the effectiveness of reinforcement learning in navigating complex optimization landscapes with strict constraints. Furthermore, we propose acceleration techniques within the OpenNeoMC framework, including model updating and cross-section usage by RAM utilization, to expedite simulation times. Our findings emphasize the importance of machine learning integration in reactor optimization and contribute to advancing methodologies for addressing intricate optimization challenges in nuclear engineering. The sources of this work are available at our GitHub repository: https://github.com/Scientific-Computing-Lab-NRCN/RLOpenNeoMC
☆ A Framework for Portrait Stylization with Skin-Tone Awareness and Nudity Identification ICASSP 2024
Portrait stylization is a challenging task involving the transformation of an input portrait image into a specific style while preserving its inherent characteristics. The recent introduction of Stable Diffusion (SD) has significantly improved the quality of outcomes in this field. However, a practical stylization framework that can effectively filter harmful input content and preserve the distinct characteristics of an input, such as skin-tone, while maintaining the quality of stylization remains lacking. These challenges have hindered the wide deployment of such a framework. To address these issues, this study proposes a portrait stylization framework that incorporates a nudity content identification module (NCIM) and a skin-tone-aware portrait stylization module (STAPSM). In experiments, NCIM showed good performance in enhancing explicit content filtering, and STAPSM accurately represented a diverse range of skin tones. Our proposed framework has been successfully deployed in practice, and it has effectively satisfied critical requirements of real-world applications.
comment: Accepted to ICASSP 2024
☆ LayoutLLM: Large Language Model Instruction Tuning for Visually Rich Document Understanding LREC
This paper proposes LayoutLLM, a more flexible document analysis method for understanding imaged documents. Visually Rich Document Understanding tasks, such as document image classification and information extraction, have gained significant attention due to their importance. Existing methods have been developed to enhance document comprehension by incorporating pre-training awareness of images, text, and layout structure. However, these methods require fine-tuning for each task and dataset, and the models are expensive to train and operate. To overcome this limitation, we propose a new LayoutLLM that integrates these with large-scale language models (LLMs). By leveraging the strengths of existing research in document image understanding and LLMs' superior language understanding capabilities, the proposed model, fine-tuned with multimodal instruction datasets, performs an understanding of document images in a single model. Our experiments demonstrate improvement over the baseline model in various document analysis tasks.
comment: LREC-COLING 2024
☆ CATSE: A Context-Aware Framework for Causal Target Sound Extraction
Target Sound Extraction (TSE) focuses on the problem of separating sources of interest, indicated by a user's cue, from the input mixture. Most existing solutions operate in an offline fashion and are not suited to the low-latency causal processing constraints imposed by applications in live-streamed content such as augmented hearing. We introduce a family of context-aware low-latency causal TSE models suitable for real-time processing. First, we explore the utility of context by providing the TSE model with oracle information about what sound classes make up the input mixture, where the objective of the model is to extract one or more sources of interest indicated by the user. Since the practical applications of oracle models are limited due to their assumptions, we introduce a composite multi-task training objective involving separation and classification losses. Our evaluation involving single- and multi-source extraction shows the benefit of using context information in the model either by means of providing full context or via the proposed multi-task training loss without the need for full context information. Specifically, we show that our proposed model outperforms size- and latency-matched Waveformer, a state-of-the-art model for real-time TSE.
comment: Submitted to EUSIPCO 2024
☆ Isotropic Gaussian Splatting for Real-Time Radiance Field Rendering
The 3D Gaussian splatting method has drawn a lot of attention, thanks to its high performance in training and high quality of the rendered image. However, it uses anisotropic Gaussian kernels to represent the scene. Although such anisotropic kernels have advantages in representing the geometry, they lead to difficulties in terms of computation, such as splitting or merging two kernels. In this paper, we propose to use isotropic Gaussian kernels to avoid such difficulties in the computation, leading to a higher performance method. The experiments confirm that the proposed method is about {\bf 100X} faster without losing the geometry representation accuracy. The proposed method can be applied in a large range applications where the radiance field is needed, such as 3D reconstruction, view synthesis, and dynamic object modeling.
☆ Dermacen Analytica: A Novel Methodology Integrating Multi-Modal Large Language Models with Machine Learning in tele-dermatology
The rise of Artificial Intelligence creates great promise in the field of medical discovery, diagnostics and patient management. However, the vast complexity of all medical domains require a more complex approach that combines machine learning algorithms, classifiers, segmentation algorithms and, lately, large language models. In this paper, we describe, implement and assess an Artificial Intelligence-empowered system and methodology aimed at assisting the diagnosis process of skin lesions and other skin conditions within the field of dermatology that aims to holistically address the diagnostic process in this domain. The workflow integrates large language, transformer-based vision models and sophisticated machine learning tools. This holistic approach achieves a nuanced interpretation of dermatological conditions that simulates and facilitates a dermatologist's workflow. We assess our proposed methodology through a thorough cross-model validation technique embedded in an evaluation pipeline that utilizes publicly available medical case studies of skin conditions and relevant images. To quantitatively score the system performance, advanced machine learning and natural language processing tools are employed which focus on similarity comparison and natural language inference. Additionally, we incorporate a human expert evaluation process based on a structured checklist to further validate our results. We implemented the proposed methodology in a system which achieved approximate (weighted) scores of 0.87 for both contextual understanding and diagnostic accuracy, demonstrating the efficacy of our approach in enhancing dermatological analysis. The proposed methodology is expected to prove useful in the development of next-generation tele-dermatology applications, enhancing remote consultation capabilities and access to care, especially in underserved areas.
Reinforcement Learning from Reflective Feedback (RLRF): Aligning and Improving LLMs via Fine-Grained Self-Reflection ACL 2024
Despite the promise of RLHF in aligning LLMs with human preferences, it often leads to superficial alignment, prioritizing stylistic changes over improving downstream performance of LLMs. Underspecified preferences could obscure directions to align the models. Lacking exploration restricts identification of desirable outputs to improve the models. To overcome these challenges, we propose a novel framework: Reinforcement Learning from Reflective Feedback (RLRF), which leverages fine-grained feedback based on detailed criteria to improve the core capabilities of LLMs. RLRF employs a self-reflection mechanism to systematically explore and refine LLM responses, then fine-tuning the models via a RL algorithm along with promising responses. Our experiments across Just-Eval, Factuality, and Mathematical Reasoning demonstrate the efficacy and transformative potential of RLRF beyond superficial surface-level adjustment.
comment: 22 pages, 5 figures, Submitted to ACL 2024
☆ A Unified Framework for Model Editing
Model editing is a growing area focused on updating the knowledge embedded within models. Among the various methodologies, ROME and MEMIT stand out as leading "locate-and-edit" model editing techniques. While MEMIT enables batched editing of memories, ROME is limited to changing one fact at a time. This paper introduces a unifying framework that brings ROME and MEMIT under a single conceptual umbrella, optimizing for the same goal, which we call the "preservation-memorization" objective. This objective aims to preserve the representations of certain selected vectors while memorizing the representations of new factual information. Specifically, ROME optimizes this objective using an equality constraint, whereas MEMIT employs a more flexible least-square constraint. In addition to making batched edits, MEMIT also edits the model at multiple layers. We disentangle the distribution of edits to multiple layers from the optimization objective of MEMIT and show that these edit-distribution algorithms should be considered separate entities worthy of their own line of research. Finally, we present EMMET - an Equality-constrained Mass Model Editing algorithm for Transformers, a new batched memory-editing algorithm. With EMMET, we present a closed form solution for the equality-constrained version of the preservation-memorization objective. We show that EMMET is able to perform batched-edits on par with MEMIT up to a batch-size of 256 and discuss the challenges in stabilizing EMMET. By articulating the "locate-and-edit" model editing algorithms under a simple conceptual framework of "preservation-memorization", we aim to bridge the gap between intuition and mathematics and hope to simplify the journey for future researchers in model editing.
SoftPatch: Unsupervised Anomaly Detection with Noisy Data
Although mainstream unsupervised anomaly detection (AD) algorithms perform well in academic datasets, their performance is limited in practical application due to the ideal experimental setting of clean training data. Training with noisy data is an inevitable problem in real-world anomaly detection but is seldom discussed. This paper considers label-level noise in image sensory anomaly detection for the first time. To solve this problem, we proposed a memory-based unsupervised AD method, SoftPatch, which efficiently denoises the data at the patch level. Noise discriminators are utilized to generate outlier scores for patch-level noise elimination before coreset construction. The scores are then stored in the memory bank to soften the anomaly detection boundary. Compared with existing methods, SoftPatch maintains a strong modeling ability of normal data and alleviates the overconfidence problem in coreset. Comprehensive experiments in various noise scenes demonstrate that SoftPatch outperforms the state-of-the-art AD methods on the MVTecAD and BTAD benchmarks and is comparable to those methods under the setting without noise.
comment: 36th Conference on Neural Information Processing Systems
☆ PeerGPT: Probing the Roles of LLM-based Peer Agents as Team Moderators and Participants in Children's Collaborative Learning
In children's collaborative learning, effective peer conversations can significantly enhance the quality of children's collaborative interactions. The integration of Large Language Model (LLM) agents into this setting explores their novel role as peers, assessing impacts as team moderators and participants. We invited two groups of participants to engage in a collaborative learning workshop, where they discussed and proposed conceptual solutions to a design problem. The peer conversation transcripts were analyzed using thematic analysis. We discovered that peer agents, while managing discussions effectively as team moderators, sometimes have their instructions disregarded. As participants, they foster children's creative thinking but may not consistently provide timely feedback. These findings highlight potential design improvements and considerations for peer agents in both roles.
comment: To appear at CHI EA '24
☆ Unsupervised Audio-Visual Segmentation with Modality Alignment
Audio-Visual Segmentation (AVS) aims to identify, at the pixel level, the object in a visual scene that produces a given sound. Current AVS methods rely on costly fine-grained annotations of mask-audio pairs, making them impractical for scalability. To address this, we introduce unsupervised AVS, eliminating the need for such expensive annotation. To tackle this more challenging problem, we propose an unsupervised learning method, named Modality Correspondence Alignment (MoCA), which seamlessly integrates off-the-shelf foundation models like DINO, SAM, and ImageBind. This approach leverages their knowledge complementarity and optimizes their joint usage for multi-modality association. Initially, we estimate positive and negative image pairs in the feature space. For pixel-level association, we introduce an audio-visual adapter and a novel pixel matching aggregation strategy within the image-level contrastive learning framework. This allows for a flexible connection between object appearance and audio signal at the pixel level, with tolerance to imaging variations such as translation and rotation. Extensive experiments on the AVSBench (single and multi-object splits) and AVSS datasets demonstrate that our MoCA outperforms strongly designed baseline methods and approaches supervised counterparts, particularly in complex scenarios with multiple auditory objects. Notably when comparing mIoU, MoCA achieves a substantial improvement over baselines in both the AVSBench (S4: +17.24%; MS3: +67.64%) and AVSS (+19.23%) audio-visual segmentation challenges.
☆ Debiasing surgeon: fantastic weights and how to find them
Nowadays an ever-growing concerning phenomenon, the emergence of algorithmic biases that can lead to unfair models, emerges. Several debiasing approaches have been proposed in the realm of deep learning, employing more or less sophisticated approaches to discourage these models from massively employing these biases. However, a question emerges: is this extra complexity really necessary? Is a vanilla-trained model already embodying some ``unbiased sub-networks'' that can be used in isolation and propose a solution without relying on the algorithmic biases? In this work, we show that such a sub-network typically exists, and can be extracted from a vanilla-trained model without requiring additional training. We further validate that such specific architecture is incapable of learning a specific bias, suggesting that there are possible architectural countermeasures to the problem of biases in deep neural networks.
☆ Quantum-activated neural reservoirs on-chip open up large hardware security models for resilient authentication
Quantum artificial intelligence is a frontier of artificial intelligence research, pioneering quantum AI-powered circuits to address problems beyond the reach of deep learning with classical architectures. This work implements a large-scale quantum-activated recurrent neural network possessing more than 3 trillion hardware nodes/cm$^2$, originating from repeatable atomic-scale nucleation dynamics in an amorphous material integrated on-chip, controlled with 0.07 nW electric power per readout channel. Compared to the best-performing reservoirs currently reported, this implementation increases the scale of the network by two orders of magnitude and reduces the power consumption by six, reaching power efficiencies in the range of the human brain, dissipating 0.2 nW/neuron. When interrogated by a classical input, the chip implements a large-scale hardware security model, enabling dictionary-free authentication secure against statistical inference attacks, including AI's present and future development, even for an adversary with a copy of all the classical components available. Experimental tests report 99.6% reliability, 100% user authentication accuracy, and an ideal 50% key uniqueness. Due to its quantum nature, the chip supports a bit density per feature size area three times higher than the best technology available, with the capacity to store more than $2^{1104}$ keys in a footprint of 1 cm$^2$. Such a quantum-powered platform could help counteract the emerging form of warfare led by the cybercrime industry in breaching authentication to target small to large-scale facilities, from private users to intelligent energy grids.
☆ StyleCineGAN: Landscape Cinemagraph Generation using a Pre-trained StyleGAN
We propose a method that can generate cinemagraphs automatically from a still landscape image using a pre-trained StyleGAN. Inspired by the success of recent unconditional video generation, we leverage a powerful pre-trained image generator to synthesize high-quality cinemagraphs. Unlike previous approaches that mainly utilize the latent space of a pre-trained StyleGAN, our approach utilizes its deep feature space for both GAN inversion and cinemagraph generation. Specifically, we propose multi-scale deep feature warping (MSDFW), which warps the intermediate features of a pre-trained StyleGAN at different resolutions. By using MSDFW, the generated cinemagraphs are of high resolution and exhibit plausible looping animation. We demonstrate the superiority of our method through user studies and quantitative comparisons with state-of-the-art cinemagraph generation methods and a video generation method that uses a pre-trained StyleGAN.
comment: Project website: https://jeolpyeoni.github.io/stylecinegan_project/
☆ OTSeg: Multi-prompt Sinkhorn Attention for Zero-Shot Semantic Segmentation
The recent success of CLIP has demonstrated promising results in zero-shot semantic segmentation by transferring muiltimodal knowledge to pixel-level classification. However, leveraging pre-trained CLIP knowledge to closely align text embeddings with pixel embeddings still has limitations in existing approaches. To address this issue, we propose OTSeg, a novel multimodal attention mechanism aimed at enhancing the potential of multiple text prompts for matching associated pixel embeddings. We first propose Multi-Prompts Sinkhorn (MPS) based on the Optimal Transport (OT) algorithm, which leads multiple text prompts to selectively focus on various semantic features within image pixels. Moreover, inspired by the success of Sinkformers in unimodal settings, we introduce the extension of MPS, called Multi-Prompts Sinkhorn Attention (MPSA), which effectively replaces cross-attention mechanisms within Transformer framework in multimodal settings. Through extensive experiments, we demonstrate that OTSeg achieves state-of-the-art (SOTA) performance with significant gains on Zero-Shot Semantic Segmentation (ZS3) tasks across three benchmark datasets.
comment: 22 pages, 7 figures
☆ Leveraging Large Language Model-based Room-Object Relationships Knowledge for Enhancing Multimodal-Input Object Goal Navigation
Object-goal navigation is a crucial engineering task for the community of embodied navigation; it involves navigating to an instance of a specified object category within unseen environments. Although extensive investigations have been conducted on both end-to-end and modular-based, data-driven approaches, fully enabling an agent to comprehend the environment through perceptual knowledge and perform object-goal navigation as efficiently as humans remains a significant challenge. Recently, large language models have shown potential in this task, thanks to their powerful capabilities for knowledge extraction and integration. In this study, we propose a data-driven, modular-based approach, trained on a dataset that incorporates common-sense knowledge of object-to-room relationships extracted from a large language model. We utilize the multi-channel Swin-Unet architecture to conduct multi-task learning incorporating with multimodal inputs. The results in the Habitat simulator demonstrate that our framework outperforms the baseline by an average of 10.6% in the efficiency metric, Success weighted by Path Length (SPL). The real-world demonstration shows that the proposed approach can efficiently conduct this task by traversing several rooms. For more details and real-world demonstrations, please check our project webpage (https://sunleyuan.github.io/ObjectNav).
comment: will soon submit to the Elsevier journal, Advanced Engineering Informatics
☆ Policy Mirror Descent with Lookahead
Policy Mirror Descent (PMD) stands as a versatile algorithmic framework encompassing several seminal policy gradient algorithms such as natural policy gradient, with connections with state-of-the-art reinforcement learning (RL) algorithms such as TRPO and PPO. PMD can be seen as a soft Policy Iteration algorithm implementing regularized 1-step greedy policy improvement. However, 1-step greedy policies might not be the best choice and recent remarkable empirical successes in RL such as AlphaGo and AlphaZero have demonstrated that greedy approaches with respect to multiple steps outperform their 1-step counterpart. In this work, we propose a new class of PMD algorithms called $h$-PMD which incorporates multi-step greedy policy improvement with lookahead depth $h$ to the PMD update rule. To solve discounted infinite horizon Markov Decision Processes with discount factor $\gamma$, we show that $h$-PMD which generalizes the standard PMD enjoys a faster dimension-free $\gamma^h$-linear convergence rate, contingent on the computation of multi-step greedy policies. We propose an inexact version of $h$-PMD where lookahead action values are estimated. Under a generative model, we establish a sample complexity for $h$-PMD which improves over prior work. Finally, we extend our result to linear function approximation to scale to large state spaces. Under suitable assumptions, our sample complexity only involves dependence on the dimension of the feature map space instead of the state space size.
☆ Deep Learning for Trajectory Data Management and Mining: A Survey and Beyond
Trajectory computing is a pivotal domain encompassing trajectory data management and mining, garnering widespread attention due to its crucial role in various practical applications such as location services, urban traffic, and public safety. Traditional methods, focusing on simplistic spatio-temporal features, face challenges of complex calculations, limited scalability, and inadequate adaptability to real-world complexities. In this paper, we present a comprehensive review of the development and recent advances in deep learning for trajectory computing (DL4Traj). We first define trajectory data and provide a brief overview of widely-used deep learning models. Systematically, we explore deep learning applications in trajectory management (pre-processing, storage, analysis, and visualization) and mining (trajectory-related forecasting, trajectory-related recommendation, trajectory classification, travel time estimation, anomaly detection, and mobility generation). Notably, we encapsulate recent advancements in Large Language Models (LLMs) that hold the potential to augment trajectory computing. Additionally, we summarize application scenarios, public datasets, and toolkits. Finally, we outline current challenges in DL4Traj research and propose future directions. Relevant papers and open-source resources have been collated and are continuously updated at: \href{https://github.com/yoshall/Awesome-Trajectory-Computing}{DL4Traj Repo}.
comment: 25 pages, 12 figures, 5 tables
☆ Evolving Benchmark Functions to Compare Evolutionary Algorithms via Genetic Programming
In this study, we use Genetic Programming (GP) to compose new optimization benchmark functions. Optimization benchmarks have the important role of showing the differences between evolutionary algorithms, making it possible for further analysis and comparisons. We show that the benchmarks generated by GP are able to differentiate algorithms better than human-made benchmark functions. The fitness measure of the GP is the Wasserstein distance of the solutions found by a pair of optimizers. Additionally, we use MAP-Elites to both enhance the search power of the GP and also illustrate how the difference between optimizers changes by various landscape features. Our approach provides a novel way to automate the design of benchmark functions and to compare evolutionary algorithms.
☆ Advancing IIoT with Over-the-Air Federated Learning: The Role of Iterative Magnitude Pruning
The industrial Internet of Things (IIoT) under Industry 4.0 heralds an era of interconnected smart devices where data-driven insights and machine learning (ML) fuse to revolutionize manufacturing. A noteworthy development in IIoT is the integration of federated learning (FL), which addresses data privacy and security among devices. FL enables edge sensors, also known as peripheral intelligence units (PIUs) to learn and adapt using their data locally, without explicit sharing of confidential data, to facilitate a collaborative yet confidential learning process. However, the lower memory footprint and computational power of PIUs inherently require deep neural network (DNN) models that have a very compact size. Model compression techniques such as pruning can be used to reduce the size of DNN models by removing unnecessary connections that have little impact on the model's performance, thus making the models more suitable for the limited resources of PIUs. Targeting the notion of compact yet robust DNN models, we propose the integration of iterative magnitude pruning (IMP) of the DNN model being trained in an over-the-air FL (OTA-FL) environment for IIoT. We provide a tutorial overview and also present a case study of the effectiveness of IMP in OTA-FL for an IIoT environment. Finally, we present future directions for enhancing and optimizing these deep compression techniques further, aiming to push the boundaries of IIoT capabilities in acquiring compact yet robust and high-performing DNN models.
comment: 6 pages, 6 figures
☆ C-TPT: Calibrated Test-Time Prompt Tuning for Vision-Language Models via Text Feature Dispersion ICLR 2024
In deep learning, test-time adaptation has gained attention as a method for model fine-tuning without the need for labeled data. A prime exemplification is the recently proposed test-time prompt tuning for large-scale vision-language models such as CLIP. Unfortunately, these prompts have been mainly developed to improve accuracy, overlooking the importance of calibration-a crucial aspect for quantifying prediction uncertainty. However, traditional calibration methods rely on substantial amounts of labeled data, making them impractical for test-time scenarios. To this end, this paper explores calibration during test-time prompt tuning by leveraging the inherent properties of CLIP. Through a series of observations, we find that the prompt choice significantly affects the calibration in CLIP, where the prompts leading to higher text feature dispersion result in better-calibrated predictions. Introducing the Average Text Feature Dispersion (ATFD), we establish its relationship with calibration error and present a novel method, Calibrated Test-time Prompt Tuning (C-TPT), for optimizing prompts during test-time with enhanced calibration. Through extensive experiments on different CLIP architectures and datasets, we show that C-TPT can effectively improve the calibration of test-time prompt tuning without needing labeled data.
comment: ICLR 2024
☆ Heuristic Algorithm-based Action Masking Reinforcement Learning (HAAM-RL) with Ensemble Inference Method
This paper presents a novel reinforcement learning (RL) approach called HAAM-RL (Heuristic Algorithm-based Action Masking Reinforcement Learning) for optimizing the color batching re-sequencing problem in automobile painting processes. The existing heuristic algorithms have limitations in adequately reflecting real-world constraints and accurately predicting logistics performance. Our methodology incorporates several key techniques including a tailored Markov Decision Process (MDP) formulation, reward setting including Potential-Based Reward Shaping, action masking using heuristic algorithms (HAAM-RL), and an ensemble inference method that combines multiple RL models. The RL agent is trained and evaluated using FlexSim, a commercial 3D simulation software, integrated with our RL MLOps platform BakingSoDA. Experimental results across 30 scenarios demonstrate that HAAM-RL with an ensemble inference method achieves a 16.25% performance improvement over the conventional heuristic algorithm, with stable and consistent results. The proposed approach exhibits superior performance and generalization capability, indicating its effectiveness in optimizing complex manufacturing processes. The study also discusses future research directions, including alternative state representations, incorporating model-based RL methods, and integrating additional real-world constraints.
comment: 7 pages, 8 figures
☆ DouRN: Improving DouZero by Residual Neural Networks
Deep reinforcement learning has made significant progress in games with imperfect information, but its performance in the card game Doudizhu (Chinese Poker/Fight the Landlord) remains unsatisfactory. Doudizhu is different from conventional games as it involves three players and combines elements of cooperation and confrontation, resulting in a large state and action space. In 2021, a Doudizhu program called DouZero\cite{zha2021douzero} surpassed previous models without prior knowledge by utilizing traditional Monte Carlo methods and multilayer perceptrons. Building on this work, our study incorporates residual networks into the model, explores different architectural designs, and conducts multi-role testing. Our findings demonstrate that this model significantly improves the winning rate within the same training time. Additionally, we introduce a call scoring system to assist the agent in deciding whether to become a landlord. With these enhancements, our model consistently outperforms the existing version of DouZero and even experienced human players. \footnote{The source code is available at \url{https://github.com/Yingchaol/Douzero_Resnet.git.}
☆ Causal knowledge engineering: A case study from COVID-19
COVID-19 appeared abruptly in early 2020, requiring a rapid response amid a context of great uncertainty. Good quality data and knowledge was initially lacking, and many early models had to be developed with causal assumptions and estimations built in to supplement limited data, often with no reliable approach for identifying, validating and documenting these causal assumptions. Our team embarked on a knowledge engineering process to develop a causal knowledge base consisting of several causal BNs for diverse aspects of COVID-19. The unique challenges of the setting lead to experiments with the elicitation approach, and what emerged was a knowledge engineering method we call Causal Knowledge Engineering (CKE). The CKE provides a structured approach for building a causal knowledge base that can support the development of a variety of application-specific models. Here we describe the CKE method, and use our COVID-19 work as a case study to provide a detailed discussion and analysis of the method.
comment: 22 pages (plus 19 pages in appendices), 9 figures, submitted for review
☆ Carbon Footprint Reduction for Sustainable Data Centers in Real-Time
As machine learning workloads significantly increase energy consumption, sustainable data centers with low carbon emissions are becoming a top priority for governments and corporations worldwide. This requires a paradigm shift in optimizing power consumption in cooling and IT loads, shifting flexible loads based on the availability of renewable energy in the power grid, and leveraging battery storage from the uninterrupted power supply in data centers, using collaborative agents. The complex association between these optimization strategies and their dependencies on variable external factors like weather and the power grid carbon intensity makes this a hard problem. Currently, a real-time controller to optimize all these goals simultaneously in a dynamic real-world setting is lacking. We propose a Data Center Carbon Footprint Reduction (DC-CFR) multi-agent Reinforcement Learning (MARL) framework that optimizes data centers for the multiple objectives of carbon footprint reduction, energy consumption, and energy cost. The results show that the DC-CFR MARL agents effectively resolved the complex interdependencies in optimizing cooling, load shifting, and energy storage in real-time for various locations under real-world dynamic weather and grid carbon intensity conditions. DC-CFR significantly outperformed the industry standard ASHRAE controller with a considerable reduction in carbon emissions (14.5%), energy usage (14.4%), and energy cost (13.7%) when evaluated over one year across multiple geographical regions.
☆ Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics
DeepFakes, which refer to AI-generated media content, have become an increasing concern due to their use as a means for disinformation. Detecting DeepFakes is currently solved with programmed machine learning algorithms. In this work, we investigate the capabilities of multimodal large language models (LLMs) in DeepFake detection. We conducted qualitative and quantitative experiments to demonstrate multimodal LLMs and show that they can expose AI-generated images through careful experimental design and prompt engineering. This is interesting, considering that LLMs are not inherently tailored for media forensic tasks, and the process does not require programming. We discuss the limitations of multimodal LLMs for these tasks and suggest possible improvements.
☆ A Roadmap Towards Automated and Regulated Robotic Systems
The rapid development of generative technology opens up possibility for higher level of automation, and artificial intelligence (AI) embodiment in robotic systems is imminent. However, due to the blackbox nature of the generative technology, the generation of the knowledge and workflow scheme is uncontrolled, especially in a dynamic environment and a complex scene. This poses challenges to regulations in safety-demanding applications such as medical scenes. We argue that the unregulated generative processes from AI is fitted for low level end tasks, but intervention in the form of manual or automated regulation should happen post-workflow-generation and pre-robotic-execution. To address this, we propose a roadmap that can lead to fully automated and regulated robotic systems. In this paradigm, the high level policies are generated as structured graph data, enabling regulatory oversight and reusability, while the code base for lower level tasks is generated by generative models. Our approach aims the transitioning from expert knowledge to regulated action, akin to the iterative processes of study, practice, scrutiny, and execution in human tasks. We identify the generative and deterministic processes in a design cycle, where generative processes serve as a text-based world simulator and the deterministic processes generate the executable system. We propose State Machine Seralization Language (SMSL) to be the conversion point between text simulator and executable workflow control. From there, we analyze the modules involved based on the current literature, and discuss human in the loop. As a roadmap, this work identifies the current possible implementation and future work. This work does not provide an implemented system but envisions to inspire the researchers working on the direction in the roadmap. We implement the SMSL and D-SFO paradigm that serve as the starting point of the roadmap.
comment: 17 pages, 9 figures
☆ The Elements of Differentiable Programming
Artificial intelligence has recently experienced remarkable advances, fueled by large models, vast datasets, accelerated hardware, and, last but not least, the transformative power of differentiable programming. This new programming paradigm enables end-to-end differentiation of complex computer programs (including those with control flows and data structures), making gradient-based optimization of program parameters possible. As an emerging paradigm, differentiable programming builds upon several areas of computer science and applied mathematics, including automatic differentiation, graphical models, optimization and statistics. This book presents a comprehensive review of the fundamental concepts useful for differentiable programming. We adopt two main perspectives, that of optimization and that of probability, with clear analogies between the two. Differentiable programming is not merely the differentiation of programs, but also the thoughtful design of programs intended for differentiation. By making programs differentiable, we inherently introduce probability distributions over their execution, providing a means to quantify the uncertainty associated with program outputs.
comment: Draft version 1
☆ RAmBLA: A Framework for Evaluating the Reliability of LLMs as Assistants in the Biomedical Domain ICLR 2024
Large Language Models (LLMs) increasingly support applications in a wide range of domains, some with potential high societal impact such as biomedicine, yet their reliability in realistic use cases is under-researched. In this work we introduce the Reliability AssesMent for Biomedical LLM Assistants (RAmBLA) framework and evaluate whether four state-of-the-art foundation LLMs can serve as reliable assistants in the biomedical domain. We identify prompt robustness, high recall, and a lack of hallucinations as necessary criteria for this use case. We design shortform tasks and tasks requiring LLM freeform responses mimicking real-world user interactions. We evaluate LLM performance using semantic similarity with a ground truth response, through an evaluator LLM.
comment: Published at ICLR 2024 Workshop on Reliable and Responsible Foundation Models
☆ Knowledge-Enhanced Recommendation with User-Centric Subgraph Network
Recommendation systems, as widely implemented nowadays on various platforms, recommend relevant items to users based on their preferences. The classical methods which rely on user-item interaction matrices has limitations, especially in scenarios where there is a lack of interaction data for new items. Knowledge graph (KG)-based recommendation systems have emerged as a promising solution. However, most KG-based methods adopt node embeddings, which do not provide personalized recommendations for different users and cannot generalize well to the new items. To address these limitations, we propose Knowledge-enhanced User-Centric subgraph Network (KUCNet), a subgraph learning approach with graph neural network (GNN) for effective recommendation. KUCNet constructs a U-I subgraph for each user-item pair that captures both the historical information of user-item interactions and the side information provided in KG. An attention-based GNN is designed to encode the U-I subgraphs for recommendation. Considering efficiency, the pruned user-centric computation graph is further introduced such that multiple U-I subgraphs can be simultaneously computed and that the size can be pruned by Personalized PageRank. Our proposed method achieves accurate, efficient, and interpretable recommendations especially for new items. Experimental results demonstrate the superiority of KUCNet over state-of-the-art KG-based and collaborative filtering (CF)-based methods.
☆ AutoRE: Document-Level Relation Extraction with Large Language Models
Large Language Models (LLMs) have demonstrated exceptional abilities in comprehending and generating text, motivating numerous researchers to utilize them for Information Extraction (IE) purposes, including Relation Extraction (RE). Nonetheless, most existing methods are predominantly designed for Sentence-level Relation Extraction (SentRE) tasks, which typically encompass a restricted set of relations and triplet facts within a single sentence. Furthermore, certain approaches resort to treating relations as candidate choices integrated into prompt templates, leading to inefficient processing and suboptimal performance when tackling Document-Level Relation Extraction (DocRE) tasks, which entail handling multiple relations and triplet facts distributed across a given document, posing distinct challenges. To overcome these limitations, we introduce AutoRE, an end-to-end DocRE model that adopts a novel RE extraction paradigm named RHF (Relation-Head-Facts). Unlike existing approaches, AutoRE does not rely on the assumption of known relation options, making it more reflective of real-world scenarios. Additionally, we have developed an easily extensible RE framework using a Parameters Efficient Fine Tuning (PEFT) algorithm (QLoRA). Our experiments on the RE-DocRED dataset showcase AutoRE's best performance, achieving state-of-the-art results, surpassing TAG by 10.03% and 9.03% respectively on the dev and test set.
comment: 11 pages
☆ Establishing a leader in a pairwise comparisons method
Abstract Like electoral systems, decision-making methods are also vulnerable to manipulation by decision-makers. The ability to effectively defend against such threats can only come from thoroughly understanding the manipulation mechanisms. In the presented article, we show two algorithms that can be used to launch a manipulation attack. They allow for equating the weights of two selected alternatives in the pairwise comparison method and, consequently, choosing a leader. The theoretical considerations are accompanied by a Monte Carlo simulation showing the relationship between the size of the PC matrix, the degree of inconsistency, and the ease of manipulation. This work is a continuation of our previous research published in the paper (Szybowski et al., 2023)
comment: 9 figures, 19 pages
Learning Quadruped Locomotion Using Differentiable Simulation
While most recent advancements in legged robot control have been driven by model-free reinforcement learning, we explore the potential of differentiable simulation. Differentiable simulation promises faster convergence and more stable training by computing low-variant first-order gradients using the robot model, but so far, its use for legged robot control has remained limited to simulation. The main challenge with differentiable simulation lies in the complex optimization landscape of robotic tasks due to discontinuities in contact-rich environments, e.g., quadruped locomotion. This work proposes a new, differentiable simulation framework to overcome these challenges. The key idea involves decoupling the complex whole-body simulation, which may exhibit discontinuities due to contact, into two separate continuous domains. Subsequently, we align the robot state resulting from the simplified model with a more precise, non-differentiable simulator to maintain sufficient simulation accuracy. Our framework enables learning quadruped walking in minutes using a single simulated robot without any parallelization. When augmented with GPU parallelization, our approach allows the quadruped robot to master diverse locomotion skills, including trot, pace, bound, and gallop, on challenging terrains in minutes. Additionally, our policy achieves robust locomotion performance in the real world zero-shot. To the best of our knowledge, this work represents the first demonstration of using differentiable simulation for controlling a real quadruped robot. This work provides several important insights into using differentiable simulations for legged locomotion in the real world.
☆ Comparing Plausibility Estimates in Base and Instruction-Tuned Large Language Models
Instruction-tuned LLMs can respond to explicit queries formulated as prompts, which greatly facilitates interaction with human users. However, prompt-based approaches might not always be able to tap into the wealth of implicit knowledge acquired by LLMs during pre-training. This paper presents a comprehensive study of ways to evaluate semantic plausibility in LLMs. We compare base and instruction-tuned LLM performance on an English sentence plausibility task via (a) explicit prompting and (b) implicit estimation via direct readout of the probabilities models assign to strings. Experiment 1 shows that, across model architectures and plausibility datasets, (i) log likelihood ($\textit{LL}$) scores are the most reliable indicator of sentence plausibility, with zero-shot prompting yielding inconsistent and typically poor results; (ii) $\textit{LL}$-based performance is still inferior to human performance; (iii) instruction-tuned models have worse $\textit{LL}$-based performance than base models. In Experiment 2, we show that $\textit{LL}$ scores across models are modulated by context in the expected way, showing high performance on three metrics of context-sensitive plausibility and providing a direct match to explicit human plausibility judgments. Overall, $\textit{LL}$ estimates remain a more reliable measure of plausibility in LLMs than direct prompting.
☆ Local Causal Discovery with Linear non-Gaussian Cyclic Models AISTATS 2024
Local causal discovery is of great practical significance, as there are often situations where the discovery of the global causal structure is unnecessary, and the interest lies solely on a single target variable. Most existing local methods utilize conditional independence relations, providing only a partially directed graph, and assume acyclicity for the ground-truth structure, even though real-world scenarios often involve cycles like feedback mechanisms. In this work, we present a general, unified local causal discovery method with linear non-Gaussian models, whether they are cyclic or acyclic. We extend the application of independent component analysis from the global context to independent subspace analysis, enabling the exact identification of the equivalent local directed structures and causal strengths from the Markov blanket of the target variable. We also propose an alternative regression-based method in the particular acyclic scenarios. Our identifiability results are empirically validated using both synthetic and real-world datasets.
comment: Appears at AISTATS 2024
☆ Crowdsourced Multilingual Speech Intelligibility Testing
With the advent of generative audio features, there is an increasing need for rapid evaluation of their impact on speech intelligibility. Beyond the existing laboratory measures, which are expensive and do not scale well, there has been comparatively little work on crowdsourced assessment of intelligibility. Standards and recommendations are yet to be defined, and publicly available multilingual test materials are lacking. In response to this challenge, we propose an approach for a crowdsourced intelligibility assessment. We detail the test design, the collection and public release of the multilingual speech data, and the results of our early experiments.
☆ The opportunities and risks of large language models in mental health
Global rates of mental health concerns are rising and there is increasing realization that existing models of mental healthcare will not adequately expand to meet the demand. With the emergence of large language models (LLMs) has come great optimism regarding their promise to create novel, large-scale solutions to support mental health. Despite their nascence, LLMs have already been applied to mental health-related tasks. In this review, we summarize the extant literature on efforts to use LLMs to provide mental health education, assessment, and intervention and highlight key opportunities for positive impact in each area. We then highlight risks associated with LLMs application to mental health and encourage adoption of strategies to mitigate these risks. The urgent need for mental health support must be balanced with responsible development, testing, and deployment of mental health LLMs. Especially critical is ensuring that mental health LLMs are fine-tuned for mental health, enhance mental health equity, adhere to ethical standards, and that people, including those with lived experience with mental health concerns, are involved in all stages from development through deployment. Prioritizing these efforts will minimize potential harms to mental health and maximize the likelihood that LLMs will positively impact mental health globally.
comment: 12 pages, 2 tables, 4 figures
☆ Deep Active Learning: A Reality Check
We conduct a comprehensive evaluation of state-of-the-art deep active learning methods. Surprisingly, under general settings, no single-model method decisively outperforms entropy-based active learning, and some even fall short of random sampling. We delve into overlooked aspects like starting budget, budget step, and pretraining's impact, revealing their significance in achieving superior results. Additionally, we extend our evaluation to other tasks, exploring the active learning effectiveness in combination with semi-supervised learning, and object detection. Our experiments provide valuable insights and concrete recommendations for future active learning studies. By uncovering the limitations of current methods and understanding the impact of different experimental settings, we aim to inspire more efficient training of deep learning models in real-world scenarios with limited annotation budgets. This work contributes to advancing active learning's efficacy in deep learning and empowers researchers to make informed decisions when applying active learning to their tasks.
☆ Planning and Acting While the Clock Ticks
Standard temporal planning assumes that planning takes place offline and then execution starts at time 0. Recently, situated temporal planning was introduced, where planning starts at time 0 and execution occurs after planning terminates. Situated temporal planning reflects a more realistic scenario where time passes during planning. However, in situated temporal planning a complete plan must be generated before any action is executed. In some problems with time pressure, timing is too tight to complete planning before the first action must be executed. For example, an autonomous car that has a truck backing towards it should probably move out of the way now and plan how to get to its destination later. In this paper, we propose a new problem setting: concurrent planning and execution, in which actions can be dispatched (executed) before planning terminates. Unlike previous work on planning and execution, we must handle wall clock deadlines that affect action applicability and goal achievement (as in situated planning) while also supporting dispatching actions before a complete plan has been found. We extend previous work on metareasoning for situated temporal planning to develop an algorithm for this new setting. Our empirical evaluation shows that when there is strong time pressure, our approach outperforms situated temporal planning.
☆ Particip-AI: A Democratic Surveying Framework for Anticipating Future AI Use Cases, Harms and Benefits
General purpose AI, such as ChatGPT, seems to have lowered the barriers for the public to use AI and harness its power. However, the governance and development of AI still remain in the hands of a few, and the pace of development is accelerating without proper assessment of risks. As a first step towards democratic governance and risk assessment of AI, we introduce Particip-AI, a framework to gather current and future AI use cases and their harms and benefits from non-expert public. Our framework allows us to study more nuanced and detailed public opinions on AI through collecting use cases, surfacing diverse harms through risk assessment under alternate scenarios (i.e., developing and not developing a use case), and illuminating tensions over AI development through making a concluding choice on its development. To showcase the promise of our framework towards guiding democratic AI, we gather responses from 295 demographically diverse participants. We find that participants' responses emphasize applications for personal life and society, contrasting with most current AI development's business focus. This shows the value of surfacing diverse harms that are complementary to expert assessments. Furthermore, we found that perceived impact of not developing use cases predicted participants' judgements of whether AI use cases should be developed, and highlighted lay users' concerns of techno-solutionism. We conclude with a discussion on how frameworks like Particip-AI can further guide democratic AI governance and regulation.
comment: 35 pages, 4 figures, 23 tables
☆ Latent Diffusion Models for Attribute-Preserving Image Anonymization
Generative techniques for image anonymization have great potential to generate datasets that protect the privacy of those depicted in the images, while achieving high data fidelity and utility. Existing methods have focused extensively on preserving facial attributes, but failed to embrace a more comprehensive perspective that considers the scene and background into the anonymization process. This paper presents, to the best of our knowledge, the first approach to image anonymization based on Latent Diffusion Models (LDMs). Every element of a scene is maintained to convey the same meaning, yet manipulated in a way that makes re-identification difficult. We propose two LDMs for this purpose: CAMOUFLaGE-Base exploits a combination of pre-trained ControlNets, and a new controlling mechanism designed to increase the distance between the real and anonymized images. CAMOFULaGE-Light is based on the Adapter technique, coupled with an encoding designed to efficiently represent the attributes of different persons in a scene. The former solution achieves superior performance on most metrics and benchmarks, while the latter cuts the inference time in half at the cost of fine-tuning a lightweight module. We show through extensive experimental comparison that the proposed method is competitive with the state-of-the-art concerning identity obfuscation whilst better preserving the original content of the image and tackling unresolved challenges that current solutions fail to address.
☆ Multi-Agent VQA: Exploring Multi-Agent Foundation Models in Zero-Shot Visual Question Answering
This work explores the zero-shot capabilities of foundation models in Visual Question Answering (VQA) tasks. We propose an adaptive multi-agent system, named Multi-Agent VQA, to overcome the limitations of foundation models in object detection and counting by using specialized agents as tools. Unlike existing approaches, our study focuses on the system's performance without fine-tuning it on specific VQA datasets, making it more practical and robust in the open world. We present preliminary experimental results under zero-shot scenarios and highlight some failure cases, offering new directions for future research.
comment: A full version of the paper will be released soon. The codes are available at https://github.com/bowen-upenn/Multi-Agent-VQA
☆ StreamingT2V: Consistent, Dynamic, and Extendable Long Video Generation from Text
Text-to-video diffusion models enable the generation of high-quality videos that follow text instructions, making it easy to create diverse and individual content. However, existing approaches mostly focus on high-quality short video generation (typically 16 or 24 frames), ending up with hard-cuts when naively extended to the case of long video synthesis. To overcome these limitations, we introduce StreamingT2V, an autoregressive approach for long video generation of 80, 240, 600, 1200 or more frames with smooth transitions. The key components are:(i) a short-term memory block called conditional attention module (CAM), which conditions the current generation on the features extracted from the previous chunk via an attentional mechanism, leading to consistent chunk transitions, (ii) a long-term memory block called appearance preservation module, which extracts high-level scene and object features from the first video chunk to prevent the model from forgetting the initial scene, and (iii) a randomized blending approach that enables to apply a video enhancer autoregressively for infinitely long videos without inconsistencies between chunks. Experiments show that StreamingT2V generates high motion amount. In contrast, all competing image-to-video methods are prone to video stagnation when applied naively in an autoregressive manner. Thus, we propose with StreamingT2V a high-quality seamless text-to-long video generator that outperforms competitors with consistency and motion. Our code will be available at: https://github.com/Picsart-AI-Research/StreamingT2V
comment: https://github.com/Picsart-AI-Research/StreamingT2V
☆ Improving Robustness to Model Inversion Attacks via Sparse Coding Architectures
Recent model inversion attack algorithms permit adversaries to reconstruct a neural network's private training data just by repeatedly querying the network and inspecting its outputs. In this work, we develop a novel network architecture that leverages sparse-coding layers to obtain superior robustness to this class of attacks. Three decades of computer science research has studied sparse coding in the context of image denoising, object recognition, and adversarial misclassification settings, but to the best of our knowledge, its connection to state-of-the-art privacy vulnerabilities remains unstudied. However, sparse coding architectures suggest an advantageous means to defend against model inversion attacks because they allow us to control the amount of irrelevant private information encoded in a network's intermediate representations in a manner that can be computed efficiently during training and that is known to have little effect on classification accuracy. Specifically, compared to networks trained with a variety of state-of-the-art defenses, our sparse-coding architectures maintain comparable or higher classification accuracy while degrading state-of-the-art training data reconstructions by factors of 1.1 to 18.3 across a variety of reconstruction quality metrics (PSNR, SSIM, FID). This performance advantage holds across 5 datasets ranging from CelebA faces to medical images and CIFAR-10, and across various state-of-the-art SGD-based and GAN-based inversion attacks, including Plug-&-Play attacks. We provide a cluster-ready PyTorch codebase to promote research and standardize defense evaluations.
comment: 32 pages, 15 Tables, and 9 Figures
♻ ☆ Emergent Dominance Hierarchies in Reinforcement Learning Agents
Modern Reinforcement Learning (RL) algorithms are able to outperform humans in a wide variety of tasks. Multi-agent reinforcement learning (MARL) settings present additional challenges, and successful cooperation in mixed-motive groups of agents depends on a delicate balancing act between individual and group objectives. Social conventions and norms, often inspired by human institutions, are used as tools for striking this balance. In this paper, we examine a fundamental, well-studied social convention that underlies cooperation in both animal and human societies: dominance hierarchies. We adapt the ethological theory of dominance hierarchies to artificial agents, borrowing the established terminology and definitions with as few amendments as possible. We demonstrate that populations of RL agents, operating without explicit programming or intrinsic rewards, can invent, learn, enforce, and transmit a dominance hierarchy to new populations. The dominance hierarchies that emerge have a similar structure to those studied in chickens, mice, fish, and other species.
♻ ☆ Intelligent Canvas: Enabling Design-Like Exploratory Visual Data Analysis with Generative AI through Rapid Prototyping, Iteration and Curation
Complex data analysis inherently seeks unexpected insights through exploratory visual analysis methods, transcending logical, step-by-step processing. However, existing interfaces such as notebooks and dashboards have limitations in exploration and comparison for visual data analysis. Addressing these limitations, we introduce a "design-like" intelligent canvas environment integrating generative AI into data analysis, offering rapid prototyping, iteration, and comparative visualization management. Our dual contributions include the integration of generative AI components into a canvas interface, and empirical findings from a user study (N=10) evaluating the effectiveness of the canvas interface.
♻ ☆ Unimodal Multi-Task Fusion for Emotional Mimicry Prediciton
In this study, we propose a methodology for the Emotional Mimicry Intensity (EMI) Estimation task within the context of the 6th Workshop and Competition on Affective Behavior Analysis in-the-wild. Our approach leverages the Wav2Vec 2.0 framework, pre-trained on a comprehensive podcast dataset, to extract a broad range of audio features encompassing both linguistic and paralinguistic elements. We enhance feature representation through a fusion technique that integrates individual features with a global mean vector, introducing global contextual insights into our analysis. Additionally, we incorporate a pre-trained valence-arousal-dominance (VAD) module from the Wav2Vec 2.0 model. Our fusion employs a Long Short-Term Memory (LSTM) architecture for efficient temporal analysis of audio data. Utilizing only the provided audio data, our approach demonstrates significant improvements over the established baseline.
♻ ☆ EasyInstruct: An Easy-to-use Instruction Processing Framework for Large Language Models
In recent years, instruction tuning has gained increasing attention and emerged as a crucial technique to enhance the capabilities of Large Language Models (LLMs). To construct high-quality instruction datasets, many instruction processing approaches have been proposed, aiming to achieve a delicate balance between data quantity and data quality. Nevertheless, due to inconsistencies that persist among various instruction processing methods, there is no standard open-source instruction processing implementation framework available for the community, which hinders practitioners from further developing and advancing. To facilitate instruction processing research and development, we present EasyInstruct, an easy-to-use instruction processing framework for LLMs, which modularizes instruction generation, selection, and prompting, while also considering their combination and interaction. EasyInstruct is publicly released and actively maintained at https://github.com/zjunlp/EasyInstruct, along with an online demo app and a demo video for quick-start, calling for broader research centered on instruction data and synthetic data.
comment: Project website: https://zjunlp.github.io/project/EasyInstruct Code: https://github.com/zjunlp/EasyInstruct Video: https://youtu.be/rfQOWYfziFo Demo: https://huggingface.co/spaces/zjunlp/EasyInstruct
♻ ☆ An explainable three dimension framework to uncover learning patterns: A unified look in variable sulci recognition
Explainable AI is crucial in medical imaging. In the challenging field of neuroscience, visual topics present a high level of complexity, particularly within three-dimensional space. The application of neuroscience, which involves identifying brain sulcal features from MRI, faces significant hurdles due to varying annotation protocols among experts and the intricate three-dimension functionality of the brain. Consequently, traditional explainability approaches fall short in effectively validating and evaluating these networks. To address this, we first present a mathematical formulation delineating various categories of explanation needs across diverse computer vision tasks, categorized into self-explanatory, semi-explanatory, non-explanatory, and new-pattern learning applications based on the reliability of the validation protocol. With respect to this mathematical formulation, we propose a 3D explainability framework aimed at validating the outputs of deep learning networks in detecting the paracingulate sulcus an essential brain anatomical feature. The framework integrates local 3D explanations, global explanations through dimensionality reduction, concatenated global explanations, and statistical shape features, unveiling new insights into pattern learning. We trained and tested two advanced 3D deep learning networks on the challenging TOP-OSLO dataset, significantly improving sulcus detection accuracy, particularly on the left hemisphere. During evaluation with diverse annotation protocols for this dataset, we highlighted the crucial role of an unbiased annotation process in achieving precise predictions and effective pattern learning within our proposed 3D framework. The proposed framework not only annotates the variable sulcus but also uncovers hidden AI knowledge, promising to advance our understanding of brain anatomy and function.
♻ ☆ Optimizing Language Augmentation for Multilingual Large Language Models: A Case Study on Korean
Large language models (LLMs) use pretraining to predict the subsequent word; however, their expansion requires significant computing resources. Numerous big tech companies and research institutes have developed multilingual LLMs (MLLMs) to meet current demands, overlooking less-resourced languages (LRLs). This study proposed three strategies to enhance the performance of LRLs based on the publicly available MLLMs. First, the MLLM vocabularies of LRLs were expanded to enhance expressiveness. Second, bilingual data were used for pretraining to align the high- and less-resourced languages. Third, a high-quality small-scale instruction dataset was constructed and instruction-tuning was performed to augment the LRL. The experiments employed the Llama2 model and Korean was used as the LRL, which was quantitatively evaluated against other developed LLMs across eight tasks. Furthermore, a qualitative assessment was performed based on human evaluation and GPT4. Experimental results showed that our proposed Bllossom model exhibited superior performance in qualitative analyses compared to previously proposed Korean monolingual models.
♻ ☆ Sequence-to-Sequence Spanish Pre-trained Language Models LREC
In recent years, significant advancements in pre-trained language models have driven the creation of numerous non-English language variants, with a particular emphasis on encoder-only and decoder-only architectures. While Spanish language models based on BERT and GPT have demonstrated proficiency in natural language understanding and generation, there remains a noticeable scarcity of encoder-decoder models explicitly designed for sequence-to-sequence tasks, which aim to map input sequences to generate output sequences conditionally. This paper breaks new ground by introducing the implementation and evaluation of renowned encoder-decoder architectures exclusively pre-trained on Spanish corpora. Specifically, we present Spanish versions of BART, T5, and BERT2BERT-style models and subject them to a comprehensive assessment across various sequence-to-sequence tasks, including summarization, question answering, split-and-rephrase, dialogue, and translation. Our findings underscore the competitive performance of all models, with the BART- and T5-based models emerging as top performers across all tasks. We have made all models publicly available to the research community to foster future explorations and advancements in Spanish NLP: https://github.com/vgaraujov/Seq2Seq-Spanish-PLMs.
comment: Accepted paper at LREC-Coling2024
♻ ☆ Effective Structured Prompting by Meta-Learning and Representative Verbalizer ICML 2023
Prompt tuning for pre-trained masked language models (MLM) has shown promising performance in natural language processing tasks with few labeled examples. It tunes a prompt for the downstream task, and a verbalizer is used to bridge the predicted token and label prediction. Due to the limited training data, prompt initialization is crucial for prompt tuning. Recently, MetaPrompting (Hou et al., 2022) uses meta-learning to learn a shared initialization for all task-specific prompts. However, a single initialization is insufficient to obtain good prompts for all tasks and samples when the tasks are complex. Moreover, MetaPrompting requires tuning the whole MLM, causing a heavy burden on computation and memory as the MLM is usually large. To address these issues, we use a prompt pool to extract more task knowledge and construct instance-dependent prompts via attention. We further propose a novel soft verbalizer (RepVerb) which constructs label embedding from feature embeddings directly. Combining meta-learning the prompt pool and RepVerb, we propose MetaPrompter for effective structured prompting. MetaPrompter is parameter-efficient as only the pool is required to be tuned. Experimental results demonstrate that MetaPrompter performs better than the recent state-of-the-arts and RepVerb outperforms existing soft verbalizers.
comment: Accepted at ICML 2023
♻ ☆ Toward a Theory of Causation for Interpreting Neural Code Models
Neural Language Models of Code, or Neural Code Models (NCMs), are rapidly progressing from research prototypes to commercial developer tools. As such, understanding the capabilities and limitations of such models is becoming critical. However, the abilities of these models are typically measured using automated metrics that often only reveal a portion of their real-world performance. While, in general, the performance of NCMs appears promising, currently much is unknown about how such models arrive at decisions. To this end, this paper introduces $do_{code}$, a post hoc interpretability method specific to NCMs that is capable of explaining model predictions. $do_{code}$ is based upon causal inference to enable programming language-oriented explanations. While the theoretical underpinnings of $do_{code}$ are extensible to exploring different model properties, we provide a concrete instantiation that aims to mitigate the impact of spurious correlations by grounding explanations of model behavior in properties of programming languages. To demonstrate the practical benefit of $do_{code}$, we illustrate the insights that our framework can provide by performing a case study on two popular deep learning architectures and ten NCMs. The results of this case study illustrate that our studied NCMs are sensitive to changes in code syntax. All our NCMs, except for the BERT-like model, statistically learn to predict tokens related to blocks of code (\eg brackets, parenthesis, semicolon) with less confounding bias as compared to other programming language constructs. These insights demonstrate the potential of $do_{code}$ as a useful method to detect and facilitate the elimination of confounding bias in NCMs.
LLM4SGG: Large Language Model for Weakly Supervised Scene Graph Generation CVPR 2024
Weakly-Supervised Scene Graph Generation (WSSGG) research has recently emerged as an alternative to the fully-supervised approach that heavily relies on costly annotations. In this regard, studies on WSSGG have utilized image captions to obtain unlocalized triplets while primarily focusing on grounding the unlocalized triplets over image regions. However, they have overlooked the two issues involved in the triplet formation process from the captions: 1) Semantic over-simplification issue arises when extracting triplets from captions, where fine-grained predicates in captions are undesirably converted into coarse-grained predicates, resulting in a long-tailed predicate distribution, and 2) Low-density scene graph issue arises when aligning the triplets in the caption with entity/predicate classes of interest, where many triplets are discarded and not used in training, leading to insufficient supervision. To tackle the two issues, we propose a new approach, i.e., Large Language Model for weakly-supervised SGG (LLM4SGG), where we mitigate the two issues by leveraging the LLM's in-depth understanding of language and reasoning ability during the extraction of triplets from captions and alignment of entity/predicate classes with target data. To further engage the LLM in these processes, we adopt the idea of Chain-of-Thought and the in-context few-shot learning strategy. To validate the effectiveness of LLM4SGG, we conduct extensive experiments on Visual Genome and GQA datasets, showing significant improvements in both Recall@K and mean Recall@K compared to the state-of-the-art WSSGG methods. A further appeal is that LLM4SGG is data-efficient, enabling effective model training with a small amount of training images.
comment: 8 pages; CVPR 2024
♻ ☆ Intrinsic Image Diffusion for Indoor Single-view Material Estimation
We present Intrinsic Image Diffusion, a generative model for appearance decomposition of indoor scenes. Given a single input view, we sample multiple possible material explanations represented as albedo, roughness, and metallic maps. Appearance decomposition poses a considerable challenge in computer vision due to the inherent ambiguity between lighting and material properties and the lack of real datasets. To address this issue, we advocate for a probabilistic formulation, where instead of attempting to directly predict the true material properties, we employ a conditional generative model to sample from the solution space. Furthermore, we show that utilizing the strong learned prior of recent diffusion models trained on large-scale real-world images can be adapted to material estimation and highly improves the generalization to real images. Our method produces significantly sharper, more consistent, and more detailed materials, outperforming state-of-the-art methods by $1.5dB$ on PSNR and by $45\%$ better FID score on albedo prediction. We demonstrate the effectiveness of our approach through experiments on both synthetic and real-world datasets.
comment: Project page: https://peter-kocsis.github.io/IntrinsicImageDiffusion/ Video: https://youtu.be/lz0meJlj5cA
♻ ☆ Point2RBox: Combine Knowledge from Synthetic Visual Patterns for End-to-end Oriented Object Detection with Single Point Supervision
With the rapidly increasing demand for oriented object detection (OOD), recent research involving weakly-supervised detectors for learning rotated box (RBox) from the horizontal box (HBox) has attracted more and more attention. In this paper, we explore a more challenging yet label-efficient setting, namely single point-supervised OOD, and present our approach called Point2RBox. Specifically, we propose to leverage two principles: 1) Synthetic pattern knowledge combination: By sampling around each labeled point on the image, we spread the object feature to synthetic visual patterns with known boxes to provide the knowledge for box regression. 2) Transform self-supervision: With a transformed input image (e.g. scaled/rotated), the output RBoxes are trained to follow the same transformation so that the network can perceive the relative size/rotation between objects. The detector is further enhanced by a few devised techniques to cope with peripheral issues, e.g. the anchor/layer assignment as the size of the object is not available in our point supervision setting. To our best knowledge, Point2RBox is the first end-to-end solution for point-supervised OOD. In particular, our method uses a lightweight paradigm, yet it achieves a competitive performance among point-supervised alternatives, 41.05%/27.62%/80.01% on DOTA/DIOR/HRSC datasets.
comment: 10 pages, 3 figures, 5 tables, code: https://github.com/yuyi1005/point2rbox-mmrotate
♻ ☆ CPSDBench: A Large Language Model Evaluation Benchmark and Baseline for Chinese Public Security Domain
Large Language Models (LLMs) have demonstrated significant potential and effectiveness across multiple application domains. To assess the performance of mainstream LLMs in public security tasks, this study aims to construct a specialized evaluation benchmark tailored to the Chinese public security domain--CPSDbench. CPSDbench integrates datasets related to public security collected from real-world scenarios, supporting a comprehensive assessment of LLMs across four key dimensions: text classification, information extraction, question answering, and text generation. Furthermore, this study introduces a set of innovative evaluation metrics designed to more precisely quantify the efficacy of LLMs in executing tasks related to public security. Through the in-depth analysis and evaluation conducted in this research, we not only enhance our understanding of the performance strengths and limitations of existing models in addressing public security issues but also provide references for the future development of more accurate and customized LLM models targeted at applications in this field.
♻ ☆ On the convergence of loss and uncertainty-based active learning algorithms
We consider the convergence rates of loss and uncertainty-based active learning algorithms under various assumptions. Firstly, we establish a set of conditions that ensure convergence rates when applied to linear classifiers and linearly separable datasets. This includes demonstrating convergence rate guarantees for loss-based sampling with various loss functions. Secondly, we introduce a framework that allows us to derive convergence rate bounds for loss-based sampling by leveraging known convergence rate bounds for stochastic gradient descent algorithms. Lastly, we propose a new algorithm that combines point sampling and stochastic Polyak's step size. We establish a condition on the sampling process, ensuring a convergence rate guarantee for this algorithm, particularly in the case of smooth convex loss functions. Our numerical results showcase the efficiency of the proposed algorithm.
♻ ☆ RiskQ: Risk-sensitive Multi-Agent Reinforcement Learning Value Factorization NeurIPS 2023
Multi-agent systems are characterized by environmental uncertainty, varying policies of agents, and partial observability, which result in significant risks. In the context of Multi-Agent Reinforcement Learning (MARL), learning coordinated and decentralized policies that are sensitive to risk is challenging. To formulate the coordination requirements in risk-sensitive MARL, we introduce the Risk-sensitive Individual-Global-Max (RIGM) principle as a generalization of the Individual-Global-Max (IGM) and Distributional IGM (DIGM) principles. This principle requires that the collection of risk-sensitive action selections of each agent should be equivalent to the risk-sensitive action selection of the central policy. Current MARL value factorization methods do not satisfy the RIGM principle for common risk metrics such as the Value at Risk (VaR) metric or distorted risk measurements. Therefore, we propose RiskQ to address this limitation, which models the joint return distribution by modeling quantiles of it as weighted quantile mixtures of per-agent return distribution utilities. RiskQ satisfies the RIGM principle for the VaR and distorted risk metrics. We show that RiskQ can obtain promising performance through extensive experiments. The source code of RiskQ is available in https://github.com/xmu-rl-3dv/RiskQ.
comment: Accepted at NeurIPS 2023
Graph Ranking Contrastive Learning: A Extremely Simple yet Efficient Method
Graph contrastive learning (GCL) has emerged as a representative graph self-supervised method, achieving significant success. The currently prevalent optimization objective for GCL is InfoNCE. Typically, it employs augmentation techniques to obtain two views, where a node in one view acts as the anchor, the corresponding node in the other view serves as the positive sample, and all other nodes are regarded as negative samples. The goal is to minimize the distance between the anchor node and positive samples and maximize the distance to negative samples. However, due to the lack of label information during training, InfoNCE inevitably treats samples from the same class as negative samples, leading to the issue of false negative samples. This can impair the learned node representations and subsequently hinder performance in downstream tasks. While numerous methods have been proposed to mitigate the impact of false negatives, they still face various challenges. For instance, while increasing the number of negative samples can dilute the impact of false negatives, it concurrently increases computational burden. Thus, we propose GraphRank, a simple yet efficient graph contrastive learning method that addresses the problem of false negative samples by redefining the concept of negative samples to a certain extent, thereby avoiding the issue of false negative samples. The effectiveness of GraphRank is empirically validated through experiments on the node, edge, and graph level tasks.
♻ ☆ Cost-Sensitive Learning to Defer to Multiple Experts with Workload Constraints
Learning to defer (L2D) aims to improve human-AI collaboration systems by learning how to defer decisions to humans when they are more likely to be correct than an ML classifier. Existing research in L2D overlooks key aspects of real-world systems that impede its practical adoption, namely: i) neglecting cost-sensitive scenarios, where type 1 and type 2 errors have different costs; ii) requiring concurrent human predictions for every instance of the training dataset and iii) not dealing with human work capacity constraints. To address these issues, we propose the deferral under cost and capacity constraints framework (DeCCaF). DeCCaF is a novel L2D approach, employing supervised learning to model the probability of human error under less restrictive data requirements (only one expert prediction per instance) and using constraint programming to globally minimize the error cost subject to workload limitations. We test DeCCaF in a series of cost-sensitive fraud detection scenarios with different teams of 9 synthetic fraud analysts, with individual work capacity constraints. The results demonstrate that our approach performs significantly better than the baselines in a wide array of scenarios, achieving an average 8.4% reduction in the misclassification cost.
♻ ☆ ROS-Causal: A ROS-based Causal Analysis Framework for Human-Robot Interaction Applications
Deploying robots in human-shared spaces requires understanding interactions among nearby agents and objects. Modelling cause-and-effect relations through causal inference aids in predicting human behaviours and anticipating robot interventions. However, a critical challenge arises as existing causal discovery methods currently lack an implementation inside the ROS ecosystem, the standard de facto in robotics, hindering effective utilisation in robotics. To address this gap, this paper introduces ROS-Causal, a ROS-based framework for onboard data collection and causal discovery in human-robot spatial interactions. An ad-hoc simulator, integrated with ROS, illustrates the approach's effectiveness, showcasing the robot onboard generation of causal models during data collection. ROS-Causal is available on GitHub: https://github.com/lcastri/roscausal.git.
comment: Accepted by the "Causal-HRI: Causal Learning for Human-Robot Interaction" workshop at the 2024 ACM/IEEE International Conference on Human-Robot Interaction (HRI)
♻ ☆ Exploring Large Language Models to Facilitate Variable Autonomy for Human-Robot Teaming
In a rapidly evolving digital landscape autonomous tools and robots are becoming commonplace. Recognizing the significance of this development, this paper explores the integration of Large Language Models (LLMs) like Generative pre-trained transformer (GPT) into human-robot teaming environments to facilitate variable autonomy through the means of verbal human-robot communication. In this paper, we introduce a novel framework for such a GPT-powered multi-robot testbed environment, based on a Unity Virtual Reality (VR) setting. This system allows users to interact with robot agents through natural language, each powered by individual GPT cores. By means of OpenAI's function calling, we bridge the gap between unstructured natural language input and structure robot actions. A user study with 12 participants explores the effectiveness of GPT-4 and, more importantly, user strategies when being given the opportunity to converse in natural language within a multi-robot environment. Our findings suggest that users may have preconceived expectations on how to converse with robots and seldom try to explore the actual language and cognitive capabilities of their robot collaborators. Still, those users who did explore where able to benefit from a much more natural flow of communication and human-like back-and-forth. We provide a set of lessons learned for future research and technical implementations of similar systems.
comment: Frontiers in Robotics and AI, Variable Autonomy for Human-Robot Teaming
♻ ☆ Knowing What LLMs DO NOT Know: A Simple Yet Effective Self-Detection Method NAACL 2024
Large Language Models (LLMs) have shown great potential in Natural Language Processing (NLP) tasks. However, recent literature reveals that LLMs generate nonfactual responses intermittently, which impedes the LLMs' reliability for further utilization. In this paper, we propose a novel self-detection method to detect which questions that a LLM does not know that are prone to generate nonfactual results. Specifically, we first diversify the textual expressions for a given question and collect the corresponding answers. Then we examine the divergencies between the generated answers to identify the questions that the model may generate falsehoods. All of the above steps can be accomplished by prompting the LLMs themselves without referring to any other external resources. We conduct comprehensive experiments and demonstrate the effectiveness of our method on recently released LLMs, e.g., Vicuna, ChatGPT, and GPT-4.
comment: Accepted by NAACL 2024
♻ ☆ SLIM: Skill Learning with Multiple Critics ICRA 2024
Self-supervised skill learning aims to acquire useful behaviors that leverage the underlying dynamics of the environment. Latent variable models, based on mutual information maximization, have been successful in this task but still struggle in the context of robotic manipulation. As it requires impacting a possibly large set of degrees of freedom composing the environment, mutual information maximization fails alone in producing useful and safe manipulation behaviors. Furthermore, tackling this by augmenting skill discovery rewards with additional rewards through a naive combination might fail to produce desired behaviors. To address this limitation, we introduce SLIM, a multi-critic learning approach for skill discovery with a particular focus on robotic manipulation. Our main insight is that utilizing multiple critics in an actor-critic framework to gracefully combine multiple reward functions leads to a significant improvement in latent-variable skill discovery for robotic manipulation while overcoming possible interference occurring among rewards which hinders convergence to useful skills. Furthermore, in the context of tabletop manipulation, we demonstrate the applicability of our novel skill discovery approach to acquire safe and efficient motor primitives in a hierarchical reinforcement learning fashion and leverage them through planning, significantly surpassing baseline approaches for skill discovery.
comment: Accepted at IEEE ICRA 2024
♻ ☆ NewsBench: Systematic Evaluation of LLMs for Writing Proficiency and Safety Adherence in Chinese Journalistic Editorial Applications
This study presents NewsBench, a novel benchmark framework developed to evaluate the capability of Large Language Models (LLMs) in Chinese Journalistic Writing Proficiency (JWP) and their Safety Adherence (SA), addressing the gap between journalistic ethics and the risks associated with AI utilization. Comprising 1,267 tasks across 5 editorial applications, 7 aspects (including safety and journalistic writing with 4 detailed facets), and spanning 24 news topics domains, NewsBench employs two GPT-4 based automatic evaluation protocols validated by human assessment. Our comprehensive analysis of 10 LLMs highlighted GPT-4 and ERNIE Bot as top performers, yet revealed a relative deficiency in journalistic ethic adherence during creative writing tasks. These findings underscore the need for enhanced ethical guidance in AI-generated journalistic content, marking a step forward in aligning AI capabilities with journalistic standards and safety considerations.
comment: 27 pages
♻ ☆ FlowerFormer: Empowering Neural Architecture Encoding using a Flow-aware Graph Transformer CVPR 2024
The success of a specific neural network architecture is closely tied to the dataset and task it tackles; there is no one-size-fits-all solution. Thus, considerable efforts have been made to quickly and accurately estimate the performances of neural architectures, without full training or evaluation, for given tasks and datasets. Neural architecture encoding has played a crucial role in the estimation, and graphbased methods, which treat an architecture as a graph, have shown prominent performance. For enhanced representation learning of neural architectures, we introduce FlowerFormer, a powerful graph transformer that incorporates the information flows within a neural architecture. FlowerFormer consists of two key components: (a) bidirectional asynchronous message passing, inspired by the flows; (b) global attention built on flow-based masking. Our extensive experiments demonstrate the superiority of FlowerFormer over existing neural encoding methods, and its effectiveness extends beyond computer vision models to include graph neural networks and auto speech recognition models. Our code is available at http://github.com/y0ngjaenius/CVPR2024_FLOWERFormer.
comment: CVPR 2024 Camera-Ready
♻ ☆ Deep Classifier Mimicry without Data Access
Access to pre-trained models has recently emerged as a standard across numerous machine learning domains. Unfortunately, access to the original data the models were trained on may not equally be granted. This makes it tremendously challenging to fine-tune, compress models, adapt continually, or to do any other type of data-driven update. We posit that original data access may however not be required. Specifically, we propose Contrastive Abductive Knowledge Extraction (CAKE), a model-agnostic knowledge distillation procedure that mimics deep classifiers without access to the original data. To this end, CAKE generates pairs of noisy synthetic samples and diffuses them contrastively toward a model's decision boundary. We empirically corroborate CAKE's effectiveness using several benchmark datasets and various architectural choices, paving the way for broad application.
comment: 11 pages main, 4 figures, 2 tables, 4 pages appendix
♻ ☆ TensorBank: Tensor Lakehouse for Foundation Model Training
Storing and streaming high dimensional data for foundation model training became a critical requirement with the rise of foundation models beyond natural language. In this paper we introduce TensorBank, a petabyte scale tensor lakehouse capable of streaming tensors from Cloud Object Store (COS) to GPU memory at wire speed based on complex relational queries. We use Hierarchical Statistical Indices (HSI) for query acceleration. Our architecture allows to directly address tensors on block level using HTTP range reads. Once in GPU memory, data can be transformed using PyTorch transforms. We provide a generic PyTorch dataset type with a corresponding dataset factory translating relational queries and requested transformations as an instance. By making use of the HSI, irrelevant blocks can be skipped without reading them as those indices contain statistics on their content at different hierarchical resolution levels. This is an opinionated architecture powered by open standards and making heavy use of open-source technology. Although, hardened for production use using geospatial-temporal data, this architecture generalizes to other use case like computer vision, computational neuroscience, biological sequence analysis and more.
♻ ☆ Detecting Sexual Content at the Sentence Level in First Millennium Latin Texts
In this study, we propose to evaluate the use of deep learning methods for semantic classification at the sentence level to accelerate the process of corpus building in the field of humanities and linguistics, a traditional and time-consuming task. We introduce a novel corpus comprising around 2500 sentences spanning from 300 BCE to 900 CE including sexual semantics (medical, erotica, etc.). We evaluate various sentence classification approaches and different input embedding layers, and show that all consistently outperform simple token-based searches. We explore the integration of idiolectal and sociolectal metadata embeddings (centuries, author, type of writing), but find that it leads to overfitting. Our results demonstrate the effectiveness of this approach, achieving high precision and true positive rates (TPR) of respectively 70.60% and 86.33% using HAN. We evaluate the impact of the dataset size on the model performances (420 instead of 2013), and show that, while our models perform worse, they still offer a high enough precision and TPR, even without MLM, respectively 69% and 51%. Given the result, we provide an analysis of the attention mechanism as a supporting added value for humanists in order to produce more data.
♻ ☆ LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models
Efficient fine-tuning is vital for adapting large language models (LLMs) to downstream tasks. However, it requires non-trivial efforts to implement these methods on different models. We present LlamaFactory, a unified framework that integrates a suite of cutting-edge efficient training methods. It allows users to flexibly customize the fine-tuning of 100+ LLMs without the need for coding through the built-in web UI LlamaBoard. We empirically validate the efficiency and effectiveness of our framework on language modeling and text generation tasks. It has been released at https://github.com/hiyouga/LLaMA-Factory and already received over 13,000 stars and 1,600 forks.
comment: 12 pages, preprint
♻ ☆ LMM-Assisted Breast Cancer Treatment Target Segmentation with Consistency Embedding
Recent advancements in Artificial Intelligence (AI) have profoundly influenced medical fields, by providing tools to reduce clinical workloads. However, most AI models are constrained to execute unimodal tasks, in stark contrast to the comprehensive approaches utilized by medical professionals. To address this, here we present RO-LMM, a multi-purpose large multimodal model (LMM) tailored for the field of radiation oncology. This model covers series of tasks within clinical workflow, adept at clinical report summarization, radiation treatment plan suggestion, and plan-guided target volume segmentation. In particular, to perform consecutive clinical tasks, we further present a novel Consistency Embedding Fine-Tuning (CEFTune) technique, which boosts LMM's robustness to noisy inputs while preserving the capability of handling clean inputs, and transform this concept into LMM-driven segmentation framework as Consistency Embedding Segmentation~(CESEG). Experimental results on multi-centre cohorts demonstrate our RO-LMM's promising performance for multiple clinical tasks with generalization capabilities.
comment: 30 pages, 16 table, 5 figures
♻ ☆ ED-NeRF: Efficient Text-Guided Editing of 3D Scene with Latent Space NeRF ICLR 2024
Recently, there has been a significant advancement in text-to-image diffusion models, leading to groundbreaking performance in 2D image generation. These advancements have been extended to 3D models, enabling the generation of novel 3D objects from textual descriptions. This has evolved into NeRF editing methods, which allow the manipulation of existing 3D objects through textual conditioning. However, existing NeRF editing techniques have faced limitations in their performance due to slow training speeds and the use of loss functions that do not adequately consider editing. To address this, here we present a novel 3D NeRF editing approach dubbed ED-NeRF by successfully embedding real-world scenes into the latent space of the latent diffusion model (LDM) through a unique refinement layer. This approach enables us to obtain a NeRF backbone that is not only faster but also more amenable to editing compared to traditional image space NeRF editing. Furthermore, we propose an improved loss function tailored for editing by migrating the delta denoising score (DDS) distillation loss, originally used in 2D image editing to the three-dimensional domain. This novel loss function surpasses the well-known score distillation sampling (SDS) loss in terms of suitability for editing purposes. Our experimental results demonstrate that ED-NeRF achieves faster editing speed while producing improved output quality compared to state-of-the-art 3D editing models.
comment: ICLR 2024; Project Page: https://jhq1234.github.io/ed-nerf.github.io/
♻ ☆ QH9: A Quantum Hamiltonian Prediction Benchmark for QM9 Molecules NeurIPS 2023
Supervised machine learning approaches have been increasingly used in accelerating electronic structure prediction as surrogates of first-principle computational methods, such as density functional theory (DFT). While numerous quantum chemistry datasets focus on chemical properties and atomic forces, the ability to achieve accurate and efficient prediction of the Hamiltonian matrix is highly desired, as it is the most important and fundamental physical quantity that determines the quantum states of physical systems and chemical properties. In this work, we generate a new Quantum Hamiltonian dataset, named as QH9, to provide precise Hamiltonian matrices for 999 or 2998 molecular dynamics trajectories and 130,831 stable molecular geometries, based on the QM9 dataset. By designing benchmark tasks with various molecules, we show that current machine learning models have the capacity to predict Hamiltonian matrices for arbitrary molecules. Both the QH9 dataset and the baseline models are provided to the community through an open-source benchmark, which can be highly valuable for developing machine learning methods and accelerating molecular and materials design for scientific and technological applications. Our benchmark is publicly available at https://github.com/divelab/AIRS/tree/main/OpenDFT/QHBench.
comment: Accepted by NeurIPS 2023, Track on Datasets and Benchmarks
♻ ☆ ANLS* -- A Universal Document Processing Metric for Generative Large Language Models
Traditionally, discriminative models have been the predominant choice for tasks like document classification and information extraction. These models make predictions that fall into a limited number of predefined classes, facilitating a binary true or false evaluation and enabling the direct calculation of metrics such as the F1 score. However, recent advancements in generative large language models (GLLMs) have prompted a shift in the field due to their enhanced zero-shot capabilities, which eliminate the need for a downstream dataset and computationally expensive fine-tuning. However, evaluating GLLMs presents a challenge as the binary true or false evaluation used for discriminative models is not applicable to the predictions made by GLLMs. This paper introduces a new metric for generative models called ANLS* for evaluating a wide variety of tasks, including information extraction and classification tasks. The ANLS* metric extends existing ANLS metrics as a drop-in-replacement and is still compatible with previously reported ANLS scores. An evaluation of 7 different datasets, 6 different GLLMs and 3 different prompting methods using the ANLS* metric is also provided, demonstrating the importance of the proposed metric. We also benchmark a novel approach to generate prompts for documents, called SFT, against other prompting techniques such as LATIN. In 27 out of 35 cases, SFT outperforms other techniques and improves the state-of-the-art, sometimes by as much as $18$ percentage points. Sources are available at https://github.com/deepopinion/anls_star_metric
♻ ☆ Distilling and Retrieving Generalizable Knowledge for Robot Manipulation via Language Corrections
Today's robot policies exhibit subpar performance when faced with the challenge of generalizing to novel environments. Human corrective feedback is a crucial form of guidance to enable such generalization. However, adapting to and learning from online human corrections is a non-trivial endeavor: not only do robots need to remember human feedback over time to retrieve the right information in new settings and reduce the intervention rate, but also they would need to be able to respond to feedback that can be arbitrary corrections about high-level human preferences to low-level adjustments to skill parameters. In this work, we present Distillation and Retrieval of Online Corrections (DROC), a large language model (LLM)-based system that can respond to arbitrary forms of language feedback, distill generalizable knowledge from corrections, and retrieve relevant past experiences based on textual and visual similarity for improving performance in novel settings. DROC is able to respond to a sequence of online language corrections that address failures in both high-level task plans and low-level skill primitives. We demonstrate that DROC effectively distills the relevant information from the sequence of online corrections in a knowledge base and retrieves that knowledge in settings with new task or object instances. DROC outperforms other techniques that directly generate robot code via LLMs by using only half of the total number of corrections needed in the first round and requires little to no corrections after two iterations. We show further results, videos, prompts and code on https://sites.google.com/stanford.edu/droc .
comment: 8 pages, 4 figures, videos and code links on website https://sites.google.com/stanford.edu/droc
♻ ☆ Exploring Neuron Interactions and Emergence in LLMs: From the Multifractal Analysis Perspective
Prior studies on the emergence in large models have primarily focused on how the functional capabilities of large language models (LLMs) scale with model size. Our research, however, transcends this traditional paradigm, aiming to deepen our understanding of the emergence within LLMs by placing a special emphasis not just on the model size but more significantly on the complex behavior of neuron interactions during the training process. By introducing the concepts of "self-organization" and "multifractal analysis," we explore how neuron interactions dynamically evolve during training, leading to "emergence," mirroring the phenomenon in natural systems where simple micro-level interactions give rise to complex macro-level behaviors. To quantitatively analyze the continuously evolving interactions among neurons in large models during training, we propose the Neuron-based Multifractal Analysis (NeuroMFA). Utilizing NeuroMFA, we conduct a comprehensive examination of the emergent behavior in LLMs through the lens of both model size and training process, paving new avenues for research into the emergence in large models.
♻ ☆ Hyper-parameter Tuning for Fair Classification without Sensitive Attribute Access
Fair machine learning methods seek to train models that balance model performance across demographic subgroups defined over sensitive attributes like race and gender. Although sensitive attributes are typically assumed to be known during training, they may not be available in practice due to privacy and other logistical concerns. Recent work has sought to train fair models without sensitive attributes on training data. However, these methods need extensive hyper-parameter tuning to achieve good results, and hence assume that sensitive attributes are known on validation data. However, this assumption too might not be practical. Here, we propose Antigone, a framework to train fair classifiers without access to sensitive attributes on either training or validation data. Instead, we generate pseudo sensitive attributes on the validation data by training a biased classifier and using the classifier's incorrectly (correctly) labeled examples as proxies for minority (majority) groups. Since fairness metrics like demographic parity, equal opportunity and subgroup accuracy can be estimated to within a proportionality constant even with noisy sensitive attribute information, we show theoretically and empirically that these proxy labels can be used to maximize fairness under average accuracy constraints. Key to our results is a principled approach to select the hyper-parameters of the biased classifier in a completely unsupervised fashion (meaning without access to ground truth sensitive attributes) that minimizes the gap between fairness estimated using noisy versus ground-truth sensitive labels.
♻ ☆ Weighted Ensemble Models Are Strong Continual Learners
In this work, we study the problem of continual learning (CL) where the goal is to learn a model on a sequence of tasks, such that the data from the previous tasks becomes unavailable while learning on the current task data. CL is essentially a balancing act between being able to learn on the new task (i.e., plasticity) and maintaining the performance on the previously learned concepts (i.e., stability). Intending to address the stability-plasticity trade-off, we propose to perform weight-ensembling of the model parameters of the previous and current tasks. This weighted-ensembled model, which we call Continual Model Averaging (or CoMA), attains high accuracy on the current task by leveraging plasticity, while not deviating too far from the previous weight configuration, ensuring stability. We also propose an improved variant of CoMA, named Continual Fisher-weighted Model Averaging (or CoFiMA), that selectively weighs each parameter in the weights ensemble by leveraging the Fisher information of the weights of the model. Both variants are conceptually simple, easy to implement, and effective in attaining state-of-the-art performance on several standard CL benchmarks. Code is available at: https://github.com/IemProg/CoFiMA.
comment: Code: https://github.com/IemProg/CoFiMA
♻ ☆ Enhancing Multimodal Cooperation via Fine-grained Modality Valuation CVPR 2024
One primary topic of multimodal learning is to jointly incorporate heterogeneous information from different modalities. However, most models often suffer from unsatisfactory multimodal cooperation, which cannot jointly utilize all modalities well. Some methods are proposed to identify and enhance the worse learnt modality, but they are often hard to provide the fine-grained observation of multimodal cooperation at sample-level with theoretical support. Hence, it is essential to reasonably observe and improve the fine-grained cooperation between modalities, especially when facing realistic scenarios where the modality discrepancy could vary across different samples. To this end, we introduce a sample-level modality valuation metric to evaluate the contribution of each modality for each sample. Via modality valuation, we observe that modality discrepancy indeed could be different at sample-level, beyond the global contribution discrepancy at dataset-level. We further analyze this issue and improve cooperation between modalities at sample-level by enhancing the discriminative ability of low-contributing modalities in a targeted manner. Overall, our methods reasonably observe the fine-grained uni-modal contribution and achieve considerable improvement. The source code and dataset are available at \url{https://github.com/GeWu-Lab/Valuate-and-Enhance-Multimodal-Cooperation}.
comment: Accepted by CVPR 2024
♻ ☆ Arcee's MergeKit: A Toolkit for Merging Large Language Models
The rapid expansion of the open-source language model landscape presents an opportunity to merge the competencies of these model checkpoints by combining their parameters. Advances in transfer learning, the process of fine-tuning pretrained models for specific tasks, has resulted in the development of vast amounts of task-specific models, typically specialized in individual tasks and unable to utilize each other's strengths. Model merging facilitates the creation of multitask models without the need for additional training, offering a promising avenue for enhancing model performance and versatility. By preserving the intrinsic capabilities of the original models, model merging addresses complex challenges in AI - including the difficulties of catastrophic forgetting and multitask learning. To support this expanding area of research, we introduce MergeKit, a comprehensive, open-source library designed to facilitate the application of model merging strategies. MergeKit offers an extensible framework to efficiently merge models on any hardware, providing utility to researchers and practitioners. To date, thousands of models have been merged by the open-source community, leading to the creation of some of the worlds most powerful open-source model checkpoints, as assessed by the Open LLM Leaderboard. The library is accessible at https://github.com/arcee-ai/MergeKit.
comment: 11 pages, 4 figures
♻ ☆ ChatGPT4PCG Competition: Character-like Level Generation for Science Birds
This paper presents the first ChatGPT4PCG Competition at the 2023 IEEE Conference on Games. The objective of this competition is for participants to create effective prompts for ChatGPT--enabling it to generate Science Birds levels with high stability and character-like qualities--fully using their creativity as well as prompt engineering skills. ChatGPT is a conversational agent developed by OpenAI. Science Birds is selected as the competition platform because designing an Angry Birds-like level is not a trivial task due to the in-game gravity; the quality of the levels is determined by their stability. To lower the entry barrier to the competition, we limit the task to the generation of capitalized English alphabetical characters. We also allow only a single prompt to be used for generating all the characters. Here, the quality of the generated levels is determined by their stability and similarity to the given characters. A sample prompt is provided to participants for their reference. An experiment is conducted to determine the effectiveness of several modified versions of this sample prompt on level stability and similarity by testing them on several characters. To the best of our knowledge, we believe that ChatGPT4PCG is the first competition of its kind and hope to inspire enthusiasm for prompt engineering in procedural content generation.
comment: This paper accepted for presentation at IEEE CoG 2023 is made available for participants of ChatGPT4PCG Competition (https://chatgpt4pcg.github.io/) and readers interested in relevant areas. In this PDF version, the affiliation symbol of Julian Togelius has been revised
♻ ☆ ComCLIP: Training-Free Compositional Image and Text Matching
Contrastive Language-Image Pretraining (CLIP) has demonstrated great zero-shot performance for matching images and text. However, it is still challenging to adapt vision-lanaguage pretrained models like CLIP to compositional image and text matching -- a more challenging image and text matching task requiring the model understanding of compositional word concepts and visual components. Towards better compositional generalization in zero-shot image and text matching, in this paper, we study the problem from a causal perspective: the erroneous semantics of individual entities are essentially confounders that cause the matching failure. Therefore, we propose a novel \textbf{\textit{training-free}} compositional CLIP model (ComCLIP). ComCLIP disentangles input images into subjects, objects, and action sub-images and composes CLIP's vision encoder and text encoder to perform evolving matching over compositional text embedding and sub-image embeddings. In this way, ComCLIP can mitigate spurious correlations introduced by the pretrained CLIP models and dynamically evaluate the importance of each component. Experiments on four compositional image-text matching datasets: SVO, ComVG, Winoground, and VL-checklist, and two general image-text retrieval datasets: Flick30K, and MSCOCO demonstrate the effectiveness of our plug-and-play method, which boosts the \textbf{\textit{zero-shot}} inference ability of CLIP, SLIP, and BLIP2 even without further training or fine-tuning. Our codes can be found at https://github.com/eric-ai-lab/ComCLIP.
♻ ☆ FourCastNeXt: Optimizing FourCastNet Training for Limited Compute
FourCastNeXt is an optimization of FourCastNet - a global machine learning weather forecasting model - that performs with a comparable level of accuracy and can be trained using around 5% of the original FourCastNet computational requirements. This technical report presents strategies for model optimization that maintain similar performance as measured by the root-mean-square error (RMSE) of the modelled variables. By providing a model with very low comparative training costs, FourCastNeXt makes Neural Earth System Modelling much more accessible to researchers looking to conduct training experiments and ablation studies. FourCastNeXt training and inference code are available at https://github.com/nci/FourCastNeXt
comment: Major revision. All prior content (text, figures, table) has been updated. Additionally, new text, tables and figures have been added. Updated title. Updated author list
♻ ☆ Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment
Ensuring alignment, which refers to making models behave in accordance with human intentions [1,2], has become a critical task before deploying large language models (LLMs) in real-world applications. For instance, OpenAI devoted six months to iteratively aligning GPT-4 before its release [3]. However, a major challenge faced by practitioners is the lack of clear guidance on evaluating whether LLM outputs align with social norms, values, and regulations. This obstacle hinders systematic iteration and deployment of LLMs. To address this issue, this paper presents a comprehensive survey of key dimensions that are crucial to consider when assessing LLM trustworthiness. The survey covers seven major categories of LLM trustworthiness: reliability, safety, fairness, resistance to misuse, explainability and reasoning, adherence to social norms, and robustness. Each major category is further divided into several sub-categories, resulting in a total of 29 sub-categories. Additionally, a subset of 8 sub-categories is selected for further investigation, where corresponding measurement studies are designed and conducted on several widely-used LLMs. The measurement results indicate that, in general, more aligned models tend to perform better in terms of overall trustworthiness. However, the effectiveness of alignment varies across the different trustworthiness categories considered. This highlights the importance of conducting more fine-grained analyses, testing, and making continuous improvements on LLM alignment. By shedding light on these key dimensions of LLM trustworthiness, this paper aims to provide valuable insights and guidance to practitioners in the field. Understanding and addressing these concerns will be crucial in achieving reliable and ethically sound deployment of LLMs in various applications.
comment: Fixed several typos
♻ ☆ VQPy: An Object-Oriented Approach to Modern Video Analytics
Video analytics is widely used in contemporary systems and services. At the forefront of video analytics are video queries that users develop to find objects of particular interest. Building upon the insight that video objects (e.g., human, animals, cars, etc.), the center of video analytics, are similar in spirit to objects modeled by traditional object-oriented languages, we propose to develop an object-oriented approach to video analytics. This approach, named VQPy, consists of a frontend$\unicode{x2015}$a Python variant with constructs that make it easy for users to express video objects and their interactions$\unicode{x2015}$as well as an extensible backend that can automatically construct and optimize pipelines based on video objects. We have implemented and open-sourced VQPy, which has been productized in Cisco as part of its DeepVision framework.
TD-MPC2: Scalable, Robust World Models for Continuous Control ICLR 2024
TD-MPC is a model-based reinforcement learning (RL) algorithm that performs local trajectory optimization in the latent space of a learned implicit (decoder-free) world model. In this work, we present TD-MPC2: a series of improvements upon the TD-MPC algorithm. We demonstrate that TD-MPC2 improves significantly over baselines across 104 online RL tasks spanning 4 diverse task domains, achieving consistently strong results with a single set of hyperparameters. We further show that agent capabilities increase with model and data size, and successfully train a single 317M parameter agent to perform 80 tasks across multiple task domains, embodiments, and action spaces. We conclude with an account of lessons, opportunities, and risks associated with large TD-MPC2 agents. Explore videos, models, data, code, and more at https://tdmpc2.com
comment: ICLR 2024. Explore videos, models, data, code, and more at https://tdmpc2.com
♻ ☆ CryCeleb: A Speaker Verification Dataset Based on Infant Cry Sounds ICASSP 2024
This paper describes the Ubenwa CryCeleb dataset - a labeled collection of infant cries - and the accompanying CryCeleb 2023 task, which is a public speaker verification challenge based on cry sounds. We released more than 6 hours of manually segmented cry sounds from 786 newborns for academic use, aiming to encourage research in infant cry analysis. The inaugural public competition attracted 59 participants, 11 of whom improved the baseline performance. The top-performing system achieved a significant improvement scoring 25.8% equal error rate, which is still far from the performance of state-of-the-art adult speaker verification systems. Therefore, we believe there is room for further research on this dataset, potentially extending beyond the verification task.
comment: ICASSP 2024
♻ ☆ Stabilizing reinforcement learning control: A modular framework for optimizing over all stable behavior
We propose a framework for the design of feedback controllers that combines the optimization-driven and model-free advantages of deep reinforcement learning with the stability guarantees provided by using the Youla-Kucera parameterization to define the search domain. Recent advances in behavioral systems allow us to construct a data-driven internal model; this enables an alternative realization of the Youla-Kucera parameterization based entirely on input-output exploration data. Perhaps of independent interest, we formulate and analyze the stability of such data-driven models in the presence of noise. The Youla-Kucera approach requires a stable "parameter" for controller design. For the training of reinforcement learning agents, the set of all stable linear operators is given explicitly through a matrix factorization approach. Moreover, a nonlinear extension is given using a neural network to express a parameterized set of stable operators, which enables seamless integration with standard deep learning libraries. Finally, we show how these ideas can also be applied to tune fixed-structure controllers.
comment: Postprint; 31 pages. arXiv admin note: text overlap with arXiv:2304.03422
♻ ☆ MacGyver: Are Large Language Models Creative Problem Solvers? NAACL 2024
We explore the creative problem-solving capabilities of modern LLMs in a novel constrained setting. To this end, we create MACGYVER, an automatically generated dataset consisting of over 1,600 real-world problems deliberately designed to trigger innovative usage of objects and necessitate out-of-the-box thinking. We then present our collection to both LLMs and humans to compare and contrast their problem-solving abilities. MACGYVER is challenging for both groups, but in unique and complementary ways. For instance, humans excel in tasks they are familiar with but struggle with domain-specific knowledge, leading to a higher variance. In contrast, LLMs, exposed to a variety of specialized knowledge, attempt broader problems but fail by proposing physically-infeasible actions. Finally, we provide a detailed error analysis of LLMs, and demonstrate the potential of enhancing their problem-solving ability with novel prompting techniques such as iterative step-wise reflection and divergent-convergent thinking. This work (1) introduces a fresh arena for intelligent agents focusing on intricate aspects of physical reasoning, planning, and unconventional thinking, which supplements the existing spectrum of machine intelligence; and (2) provides insight into the constrained problem-solving capabilities of both humans and AI.
comment: NAACL 2024
♻ ☆ Tur[k]ingBench: A Challenge Benchmark for Web Agents
Recent chatbots have demonstrated impressive ability to understand and communicate in raw-text form. However, there is more to the world than raw text. For example, humans spend long hours of their time on web pages, where text is intertwined with other modalities and tasks are accomplished in the form of various complex interactions. Can state-of-the-art multi-modal models generalize to such complex domains? To address this question, we introduce TurkingBench, a benchmark of tasks formulated as web pages containing textual instructions with multi-modal context. Unlike existing work which employs artificially synthesized web pages, here we use natural HTML pages that were originally designed for crowdsourcing workers for various annotation purposes. The HTML instructions of each task are also instantiated with various values (obtained from the crowdsourcing tasks) to form new instances of the task. This benchmark contains 32.2K instances distributed across 158 tasks. Additionally, to facilitate the evaluation on TurkingBench, we develop an evaluation framework that connects the responses of chatbots to modifications on web pages (modifying a text box, checking a radio, etc.). We evaluate the performance of state-of-the-art models, including language-only, vision-only, and layout-only models, and their combinations, on this benchmark. Our findings reveal that these models perform significantly better than random chance, yet considerable room exists for improvement. We hope this benchmark will help facilitate the evaluation and development of web-based agents.
♻ ☆ Incentivizing News Consumption on Social Media Platforms Using Large Language Models and Realistic Bot Accounts
Polarization, declining trust, and wavering support for democratic norms are pressing threats to U.S. democracy. Exposure to verified and quality news may lower individual susceptibility to these threats and make citizens more resilient to misinformation, populism, and hyperpartisan rhetoric. This project examines how to enhance users' exposure to and engagement with verified and ideologically balanced news in an ecologically valid setting. We rely on a large-scale two-week long field experiment (from 1/19/2023 to 2/3/2023) on 28,457 Twitter users. We created 28 bots utilizing GPT-2 that replied to users tweeting about sports, entertainment, or lifestyle with a contextual reply containing two hardcoded elements: a URL to the topic-relevant section of quality news organization and an encouragement to follow its Twitter account. To further test differential effects by gender of the bots, treated users were randomly assigned to receive responses by bots presented as female or male. We examine whether our over-time intervention enhances the following of news media organization, the sharing and the liking of news content and the tweeting about politics and the liking of political content. We find that the treated users followed more news accounts and the users in the female bot treatment were more likely to like news content than the control. Most of these results, however, were small in magnitude and confined to the already politically interested Twitter users, as indicated by their pre-treatment tweeting about politics. These findings have implications for social media and news organizations, and also offer direction for future work on how Large Language Models and other computational interventions can effectively enhance individual on-platform engagement with quality news and public affairs.
♻ ☆ TeleMoMa: A Modular and Versatile Teleoperation System for Mobile Manipulation
A critical bottleneck limiting imitation learning in robotics is the lack of data. This problem is more severe in mobile manipulation, where collecting demonstrations is harder than in stationary manipulation due to the lack of available and easy-to-use teleoperation interfaces. In this work, we demonstrate TeleMoMa, a general and modular interface for whole-body teleoperation of mobile manipulators. TeleMoMa unifies multiple human interfaces including RGB and depth cameras, virtual reality controllers, keyboard, joysticks, etc., and any combination thereof. In its more accessible version, TeleMoMa works using simply vision (e.g., an RGB-D camera), lowering the entry bar for humans to provide mobile manipulation demonstrations. We demonstrate the versatility of TeleMoMa by teleoperating several existing mobile manipulators - PAL Tiago++, Toyota HSR, and Fetch - in simulation and the real world. We demonstrate the quality of the demonstrations collected with TeleMoMa by training imitation learning policies for mobile manipulation tasks involving synchronized whole-body motion. Finally, we also show that TeleMoMa's teleoperation channel enables teleoperation on site, looking at the robot, or remote, sending commands and observations through a computer network, and perform user studies to evaluate how easy it is for novice users to learn to collect demonstrations with different combinations of human interfaces enabled by our system. We hope TeleMoMa becomes a helpful tool for the community enabling researchers to collect whole-body mobile manipulation demonstrations. For more information and video results, https://robin-lab.cs.utexas.edu/telemoma-web.
♻ ☆ FERGI: Automatic Annotation of User Preferences for Text-to-Image Generation from Spontaneous Facial Expression Reaction
Researchers have proposed to use data of human preference feedback to fine-tune text-to-image generative models. However, the scalability of human feedback collection has been limited by its reliance on manual annotation. Therefore, we develop and test a method to automatically annotate user preferences from their spontaneous facial expression reaction to the generated images. We collect a dataset of Facial Expression Reaction to Generated Images (FERGI) and show that the activations of multiple facial action units (AUs) are highly correlated with user evaluations of the generated images. Specifically, AU4 (brow lowerer) is reflective of negative evaluations of the generated image whereas AU12 (lip corner puller) is reflective of positive evaluations. These can be useful in two ways. Firstly, we can automatically annotate user preferences between image pairs with substantial difference in these AU responses with an accuracy significantly outperforming state-of-the-art scoring models. Secondly, directly integrating the AU responses with the scoring models improves their consistency with human preferences. Finally, this method of automatic annotation with facial expression analysis can be potentially generalized to other generation tasks. The code is available at https://github.com/ShuangquanFeng/FERGI, and the dataset is also available at the same link for research purposes.
♻ ☆ The Model Openness Framework: Promoting Completeness and Openness for Reproducibility, Transparency and Usability in AI
Generative AI (GAI) offers unprecedented possibilities but its commercialization has raised concerns about transparency, reproducibility, bias, and safety. Many "open-source" GAI models lack the necessary components for full understanding and reproduction, and some use restrictive licenses, a practice known as "openwashing." We propose the Model Openness Framework (MOF), a ranked classification system that rates machine learning models based on their completeness and openness, following principles of open science, open source, open data, and open access. The MOF requires specific components of the model development lifecycle to be included and released under appropriate open licenses. This framework aims to prevent misrepresentation of models claiming to be open, guide researchers and developers in providing all model components under permissive licenses, and help companies, academia, and hobbyists identify models that can be safely adopted without restrictions. Wide adoption of the MOF will foster a more open AI ecosystem, accelerating research, innovation, and adoption.
comment: 45 pages
Optimization and Control 47
☆ Rescue Craft Allocation in Tidal Waters of the North and Baltic Sea
This paper aims to improve the average response time for naval accidents in the North and Baltic Sea. To do this we optimize the strategic distribution of the vessel fleet used by the Deutsche Gesellschaft zur Rettung Schiffbr\"uchiger (German Maritime Search and Rescue Service) (DGzRS) across several home stations. Based on these locations, in case of an incoming distress call the vessel with the lowest response time is dispatched. A particularity of the region considered is the fact that due to low tide, at predictable times some vessels and stations are not operational. In our work, we build a corresponding mathematical model for the allocation of rescue crafts to multiple stations. Thereafter, we show that the problem is NP-hard. Next, we provide an Integer Programming (IP) formulation. Finally, we propose several methods of simplifying the model and do a case study to compare their effectiveness. For this, we generate test instances based on real-world data.
comment: Code URL: https://github.com/veni-vidi-code/RCA
☆ A Joint Optimization Approach for Power-Efficient Heterogeneous OFDMA Radio Access Networks
Heterogeneous networks have emerged as a popular solution for accommodating the growing number of connected devices and increasing traffic demands in cellular networks. While offering broader coverage, higher capacity, and lower latency, the escalating energy consumption poses sustainability challenges. In this paper a novel optimization approach for OFDMA heterogeneous networks is proposed to minimize transmission power while respecting individual users throughput constraints. The problem is formulated as a mixed integer geometric program, and optimizes at once multiple system variables such as user association, working bandwidth, and base stations transmission powers. Crucially, the proposed approach becomes a convex optimization problem when user-base station associations are provided. Evaluations in multiple realistic scenarios from the production mobile network of a major European operator and based on precise channel gains and throughput requirements from measured data validate the effectiveness of the proposed approach. Overall, our original solution paves the road for greener connectivity by reducing the energy footprint of heterogeneous mobile networks, hence fostering more sustainable communication systems.
☆ Quantitative Indicators for Strength of Inequalities with Respect to a Polyhedron, Part II: Applications and Computational Evidence
The first paper explored two strength indicators (extreme point ratio (EPR) and centroid distance (CD)), both of which predict that subtours of the spanning tree in hypergraph polytope STHGP having large cardinality are significantly stronger than the corresponding subtours of small cardinality. In this second paper, we exploit this previously unknown property algorithmically within GeoSteiner, presenting strong computational evidence that the EPR and CD indicators are highly predictive of actual computational strength, at least for subtour inequalities of STHGP. The GeoSteiner package for geometric Steiner trees in the plane uses STHGP to solve the Full Steiner Tree (FST) concatenation problem. The separation algorithms in previous versions of GeoSteiner find violated subtours of only relatively small cardinality, and we examine the underlying algorithmic causes. We present improvements in GeoSteiner specifically designed to strengthen violated subtours by augmentation (instead of reduction), yielding violated subtours of large cardinality. Across all distance metrics and instance classes studied, the computational results are remarkable -- culminating with an optimal solution of a 1,000,000 terminal random Euclidean instance. The conclusion is that the EPR and CD strength indicators presented in the first paper have strong predictive power regarding actual computational strength (at least regarding STHGP subtours). The ability to accurately measure the strength of inequalities has numerous applications of great importance, both in theory and practice.
comment: 112 pages, 18 figures
☆ First-Order Methods for Linear Programming
Linear programming is the seminal optimization problem that has spawned and grown into today's rich and diverse optimization modeling and algorithmic landscape. This article provides an overview of the recent development of first-order methods for solving large-scale linear programming.
☆ Quantitative Indicators for Strength of Inequalities with Respect to a Polyhedron, Part I: Theory
We study strength of inequalities used in mixed-integer programming, and in branch-and-cut algorithms that solve such problems. Strength is an ethereal property lacking good formal definition, but crucial for computational speed. We review several quantitative indicators proposed in the literature we claim provide a measure of the relative strength of inequalities with respect to a given polyhedron. We evaluate two of these indicators (extreme point ratio (EPR) and centroid distance (CD)) on various facet classes for both the traveling salesman polytope TSP, and spanning tree in hypergraph polytope STHGP, obtaining closed-forms for EPR and CD on each facet class. Within each facet class, the two indicators yield strikingly similar strength rankings, with excellent agreement on which facets are strongest and which are weakest. Both indicators corroborate all known computational experience with both polytopes. The indicators also reveal previously unknown properties of STHGP subtours. We also evaluate EPR and CD for the subtour inequalities of the spanning tree in graphs polytope STGP, obtaining surprising and unexpected results that (at least for STGP and STHGP subtours) lead us to believe EPR to be a more accurate estimate of strength than CD. Applications include: comparing the relative strength of different classes of inequalities; design of rapidly-converging separation algorithms; design or justification for constraint strengthening procedures. The companion paper exploits one of the newly revealed properties of STHGP subtours in GeoSteiner, with detailed computational results. Across all distance metrics and instances studied, these results are remarkable -- culminating with an optimal solution of a 1,000,000 terminal random Euclidean instance. This confirms these indicators to be highly predictive and strongly correlated with actual computational strength.
comment: 74 pages, 47 figures
☆ Modeling and optimization for arrays of water turbine OWC devices
Wave energy conversion is emerging as a promising technology for generating energy from renewable sources. Large-scale implementation of this technology requires the installation of parks of devices. We study the problem of optimizing the park layout and control for wave energy converters of the oscillating water column type. As a test case, we consider a device with a semi-submerged chamber and a Wells turbine working in the liquid phase. First, a novel model based on a nonlinear ordinary differential equation is derived to describe the behavior of the water column and used to estimate the power matrix of an isolated device. Then, its linearization is derived in order to enable the fast simulation of large parks with a high number of devices. The choice of the hydrodynamic model allows obtaining the gradient of the power with respect to the positions through an adjoint approach, making it especially convenient for optimization. We consider in particular the case of interaction with the piles of a floating wind energy plant. The results from the developed computational framework allow us to draw interesting conclusions that are useful when designing the layout of a park. In particular, we observe that interaction effects can be significant even in parks made up of devices of small size, which would exhibit negligible diffraction and radiation properties in isolated conditions, if the number of devices is large enough. Moreover, results show that wave reflection from the piles of an offshore platform can have positive effects on energy production.
☆ CBX: Python and Julia packages for consensus-based interacting particle methods
We introduce CBXPy and ConsensusBasedX.jl, Python and Julia implementations of consensus-based interacting particle systems (CBX), which generalise consensus-based optimization methods (CBO) for global, derivative-free optimisation. The raison d'\^etre of our libraries is twofold: on the one hand, to offer high-performance implementations of CBX methods that the community can use directly, while on the other, providing a general interface that can accommodate and be extended to further variations of the CBX family. Python and Julia were selected as the leading high-level languages in terms of usage and performance, as well as their popularity among the scientific computing community. Both libraries have been developed with a common ethos, ensuring a similar API and core functionality, while leveraging the strengths of each language and writing idiomatic code.
comment: 7 pages, 3 figures
☆ Spectral Methods for Quantum Optimal Control: Artificial Boundary Conditions
The problem of quantum state preparation is one of the main challenges in achieving the quantum advantage. Furthermore, classically, for multi-level problems, our ability to solve the corresponding quantum optimal control problems is rather limited. The ability of the latter to feed into the former may result in significant progress in quantum computing. To address this challenge, we propose a formulation of quantum optimal control that makes use of artificial boundary conditions for the Schr\"odinger equation in combination with spectral methods. The resulting formulations are well suited for investigating periodic potentials and lend themselves to direct numerical treatment using conventional methods for bounded domains.
comment: 10 pages plus references
☆ Breaking consensus in kinetic opinion formation models on graphons
In this work we propose and investigate a strategy to prevent consensus in kinetic models for opinion formation. We consider a large interacting agent system, and assume that agent interactions are driven by compromise as well as self-thinking dynamics and also modulated by an underlying static social network. This network structure is included using so-called graphons, which modulate the interaction frequency in the corresponding kinetic formulation. We then derive the corresponding limiting Fokker Planck equation, and analyze its large time behavior. This microscopic setting serves as a starting point for the proposed control strategy, which steers agents away from mean opinion and is characterised by a suitable penalization depending on the properties of the graphon. We show that this minimalist approach is very effective by analyzing the quasi-stationary solutions mean-field model in a plurality of graphon structures. Several numerical experiments are also provided the show the effectiveness of the approach in preventing the formation of consensus steering the system towards a declustered state.
☆ Task-optimal data-driven surrogate models for eNMPC via differentiable simulation and optimization
We present a method for end-to-end learning of Koopman surrogate models for optimal performance in control. In contrast to previous contributions that employ standard reinforcement learning (RL) algorithms, we use a training algorithm that exploits the potential differentiability of environments based on mechanistic simulation models. We evaluate the performance of our method by comparing it to that of other controller type and training algorithm combinations on a literature known eNMPC case study. Our method exhibits superior performance on this problem, thereby constituting a promising avenue towards more capable controllers that employ dynamic surrogate models.
comment: 6 pages, 4 figures, 1 table
☆ Regularized Adaptive Momentum Dual Averaging with an Efficient Inexact Subproblem Solver for Training Structured Neural Network
We propose a Regularized Adaptive Momentum Dual Averaging (RAMDA) algorithm for training structured neural networks. Similar to existing regularized adaptive methods, the subproblem for computing the update direction of RAMDA involves a nonsmooth regularizer and a diagonal preconditioner, and therefore does not possess a closed-form solution in general. We thus also carefully devise an implementable inexactness condition that retains convergence guarantees similar to the exact versions, and propose a companion efficient solver for the subproblems of both RAMDA and existing methods to make them practically feasible. We leverage the theory of manifold identification in variational analysis to show that, even in the presence of such inexactness, the iterates of RAMDA attain the ideal structure induced by the regularizer at the stationary point of asymptotic convergence. This structure is locally optimal near the point of convergence, so RAMDA is guaranteed to obtain the best structure possible among all methods converging to the same point, making it the first regularized adaptive method outputting models that possess outstanding predictive performance while being (locally) optimally structured. Extensive numerical experiments in large-scale modern computer vision, language modeling, and speech tasks show that the proposed RAMDA is efficient and consistently outperforms state of the art for training structured neural network. Implementation of our algorithm is available at http://www.github.com/ismoptgroup/RAMDA/.
☆ Infinite horizon McKean-Vlasov FBSDEs and applications to mean field control problems
In this paper, we study a class of infinite horizon fully coupled McKean-Vlasov forward-backward stochastic differential equations (FBSDEs). We propose a generalized monotonicity condition involving two flexible functions. Under this condition, we establish the well-posedness results for infinite horizon McKean-Vlasov FBSDEs by the method of continuation, including the unique solvability, an estimate of the solution, and the related continuous dependence property of the solution on the coefficients. Based on the solvability result, we study an infinite horizon mean field control problem. Moreover, by choosing appropriate form of the flexible functions, we can eliminate the different phenomenon between the linear-quadratic (LQ) problems on infinite horizon and finite horizon proposed in Wei and Yu (SIAM J. Control Optim. 59: 2594-2623, 2021).
☆ Minimal covariance realization and system identification algorithm for a class of stochastic linear switched systems with i.i.d. switching
In this paper, we work with Linear Switched Systems (LSS). We show the existence of minimality in innovation form for such systems. We also present a realization algorithm to compute a minimal LSS in innovation form, by calculating the covariances of the state-pace matrices. Finally, a system identification algorithm statically consistent is presented. The later, based on the realization algorithm, uses the collected data in order to compute the covariances of the inputs and outputs.
☆ On unifying control barrier and Lyapunov functions using QP and Sontag's formula with an application to tumor dynamics
A common tool in system theory for formulating control laws that achieve local asymptotic stability are Control Lyapunov functions (CLFs), while Control Barrier functions (CBFs) are typically employed to enforce safety constraints. Combining these two types of functions is of interest, because it leads to stabilizing controllers with safety guarantees. A common approach to merge CLFs and CBFs is to solve an optimization problem where both CLF and CBF inequalities are imposed as constraints. In this paper, we show via an example from the literature that this approach can lead to undesirable behavior (i.e., slow convergence and oscillating inputs). Then, we propose a novel cost function that penalizes the deviation from Sontag's formula by using a state-dependent weighting matrix. We show that by minimizing the developed cost function subject to a CBF constraint, local asymptotic stability is obtained with an explicit domain of attraction, without using a CLF constraint. To deal with vanishing properties of the weight matrix as the state approaches the equilibrium, we introduce a hybrid continuous control law that recovers Sontag's formula locally. The effectiveness of the developed hybrid stabilizing control law based on CLFs and CBFs is illustrated in stabilization of a 3D tumor model, subject to physiological constraints (i.e., all states must be positive), which yields useful insights into optimal cancer treatment design.
☆ Multi Methods of Matrix Analysis Use for Control and Optimization system in Control Engineering
Matrix analysis plays a crucial role in the field of control engineering, providing a powerful mathematical framework for the analysis and design of control systems. This research report explores various applications of matrix analysis in control engineering, focusing on its contributions to system modeling, stability analysis, controllablity, observability, and optimization. The report also discusses specific examples and case studies to illustrate the practical significance of matrix analysis in addressing real-world control engineering challenges Analyze controllability. Informally, a system is controllable if we can construct a set of inputs that will drive the system to any given state. Analyze observability. Informally, observability means that by controlling the inputs and watching the outputs of a system we can determine what the states were. Optimal Control is a control method that aims to find the optimal control input to achieve the best performance of the system under certain constraints. This performance index can be the system output, energy consumption, time, etc.
comment: 8 pages
Optimal Scheduling of Uplink-Downlink Networked Control Systems with Energy Harvesting Sensor
In this work, we consider a wireless networked control system (WNCS) consisting of a plant, a battery-operated sensor, a controller, and an actuator. The battery in the sensor harvests energy from the environment. The sensor then uses this energy for packet transmissions. There are two types of wireless communication channels, (i) sensor--controller channel (also called uplink channel), and (ii) controller--actuator channel (also called downlink channel). The controller is \emph{half-duplex}, and this prevents it from simultaneously receiving an update from the sensor, and also transmitting a control packet to the actuator. Though frequent transmissions via uplink channel improve controller's estimate of the plant state, but this also reduces the timely control of the plant. Hence, in order to strike a balance between these two, we consider the problem of designing an optimal scheduling policy that minimizes the expected cumulative infinite horizon discounted cost, where the instantaneous cost is equal to the square of the plant state. At each time $t$, the scheduler at the sensor has to decide whether it should activate the uplink channel, or downlink. We pose this dynamic optimization problem as a Markov decision process (MDP), in which the state at time $t$ is composed of (i) the plant state $x(t)$, (ii) the age of the data packet available at the controller, denoted by $\tau(t)$, (iii) a binary variable $y(t)$ which indicates the availability of a control packet at the controller, and (iv) the energy level of the battery at the sensor $b(t)$. We show that there exists an optimal scheduling policy that exhibits a threshold structure, meaning that for each time $t$, if there is a control packet available with the controller, then the sensor activates the downlink channel in case $|x(t)|$ exceeds a threshold $x\ust(\tau(t),b(t))$.
☆ Conservative Linear Envelopes for High-Dimensional, Hamilton-Jacobi Reachability for Nonlinear Systems via the Hopf Formula
Hamilton-Jacobi reachability (HJR) analysis provides a value function that encodes (1) the set of states from which a nonlinear system with bounded control inputs can reach a goal (or avoid a failure set) despite any bounded disturbance, and (2) the corresponding optimal control policy to reach (or avoid). Though powerful, traditional methods for HJR rely on dynamic programming and suffer from exponential computation growth with respect to state dimension. The recently favored Hopf formula mitigates this ''curse of dimensionality'' by providing an efficient and space-parallelizable approach for solving the reachability problem. However, the Hopf formula can only be applied to linear time-varying systems. To overcome this limitation, we show that the error between a nonlinear system and a linear model can be transformed into an adversarial bounded artificial disturbance, making an envelope of the true value. One may then solve the dimension-robust Hopf formula for a linear game with this ''antagonistic error" to perform guaranteed conservative reachability analysis and control synthesis of nonlinear systems; this can be done for problem formulations in which no other HJR method is both computationally feasible and guaranteed. In addition, we offer several technical methods for reducing conservativeness in the analysis. We demonstrate the theory by solving the safe linear envelope in the controlled Van der Pol system, where the true reachable set may be observed, and by solving a 5 agent (15D) pursuit-evasion game with Dubins cars.
☆ Robust Locomotion via Zero-order Stochastic Nonlinear Model Predictive Control with Guard Saltation Matrix
This paper presents a stochastic/robust nonlinear model predictive control (NMPC) to enhance the robustness of legged locomotion against contact uncertainties. We integrate the contact uncertainties into the covariance propagation of stochastic/robust NMPC framework by leveraging the guard saltation matrix and an extended Kalman filter-like covariance update. We achieve fast stochastic/robust NMPC computation by utilizing the zero-order stochastic/robust NMPC algorithm with additional improvements in computational efficiency concerning the feedback gains. We conducted numerical experiments and demonstrate that the proposed method can accurately forecast future state covariance and generate trajectories that satisfies constraints even in the presence of the contact uncertainties. Hardware experiments on the perceptive locomotion of a wheeled-legged robot were also carried out, validating the feasibility of the proposed method in a real-world system with limited on-board computation.
comment: 8 pages, 8 figures
Reinforcement Learning Design for Quickest Change Detection
The field of quickest change detection (QCD) concerns design and analysis of algorithms to estimate in real time the time at which an important event takes place, and identify properties of the post-change behavior. It is shown in this paper that approaches based on reinforcement learning (RL) can be adapted based on any "surrogate information state" that is adapted to the observations. Hence we are left to choose both the surrogate information state process and the algorithm. For the former, it is argued that there are many choices available, based on a rich theory of asymptotic statistics for QCD. Two approaches to RL design are considered: (i) Stochastic gradient descent based on an actor-critic formulation. Theory is largely complete for this approach: the algorithm is unbiased, and will converge to a local minimum. However, it is shown that variance of stochastic gradients can be very large, necessitating the need for commensurately long run times; (ii) Q-learning algorithms based on a version of the projected Bellman equation. It is shown that the algorithm is stable, in the sense of bounded sample paths, and that a solution to the projected Bellman equation exists under mild conditions. Numerical experiments illustrate these findings, and provide a roadmap for algorithm design in more general settings.
☆ Addressing complex boundary conditions of miscible flow and transport in two and three dimensions with application to optimal control
We investigate complex boundary conditions of the miscible displacement system in two and three space dimensions with the commonly-used Bear-Scheidegger diffusion-dispersion tensor, which describes, e.g., the porous medium flow processes in petroleum reservoir simulation or groundwater contaminant transport. Specifically, we incorporate the no-flux boundary condition for the Darcy velocity to prove that the general no-flux boundary condition for the transport equation is equivalent to the normal derivative boundary condition of the concentration, based on which we further prove several complex boundary conditions by the Bear-Scheidegger tensor and its derivative. The derived boundary conditions not only provide new insights and distinct properties of the Bear-Scheidegger diffusion-dispersion tensor, but accommodate the coupling and the nonlinearity of the miscible displacement system and the Bear-Scheidegger tensor in deriving the first-order optimality condition of the corresponding optimal control problem for practical application.
☆ Automatic Outlier Rectification via Optimal Transport
In this paper, we propose a novel conceptual framework to detect outliers using optimal transport with a concave cost function. Conventional outlier detection approaches typically use a two-stage procedure: first, outliers are detected and removed, and then estimation is performed on the cleaned data. However, this approach does not inform outlier removal with the estimation task, leaving room for improvement. To address this limitation, we propose an automatic outlier rectification mechanism that integrates rectification and estimation within a joint optimization framework. We take the first step to utilize an optimal transport distance with a concave cost function to construct a rectification set in the space of probability distributions. Then, we select the best distribution within the rectification set to perform the estimation task. Notably, the concave cost function we introduced in this paper is the key to making our estimator effectively identify the outlier during the optimization process. We discuss the fundamental differences between our estimator and optimal transport-based distributionally robust optimization estimator. finally, we demonstrate the effectiveness and superiority of our approach over conventional approaches in extensive simulation and empirical analyses for mean estimation, least absolute regression, and the fitting of option implied volatility surfaces.
☆ Joint Planning of Charging Stations and Power Systems for Heavy-Duty Drayage Trucks
As global concerns about climate change intensify, the transition towards zero-emission freight is becoming increasingly vital. Drayage is an important segment of the freight system, typically involving the transport of goods from seaports or intermodal terminals to nearby warehouses. This sector significantly contributes to not only greenhouse gas emissions, but also pollution in densely populated areas. This study presents a holistic optimization model designed for an efficient transition to zero-emission drayage, offering cost-effective strategies for the coordinated investment planning for power systems, charging infrastructure, and electric drayage trucks. The model is validated in the Greater Los Angeles area, where regulatory goals are among the most ambitious. Furthermore, the model's design allows for easy adaptation to other regions. By focusing on drayage trucks, this study also paves the way for future research into other freight categories, establishing a foundation for a more extensive exploration in this field.
comment: 34 pages, 10 figures
☆ Non-Convex Robust Hypothesis Testing using Sinkhorn Uncertainty Sets
We present a new framework to address the non-convex robust hypothesis testing problem, wherein the goal is to seek the optimal detector that minimizes the maximum of worst-case type-I and type-II risk functions. The distributional uncertainty sets are constructed to center around the empirical distribution derived from samples based on Sinkhorn discrepancy. Given that the objective involves non-convex, non-smooth probabilistic functions that are often intractable to optimize, existing methods resort to approximations rather than exact solutions. To tackle the challenge, we introduce an exact mixed-integer exponential conic reformulation of the problem, which can be solved into a global optimum with a moderate amount of input data. Subsequently, we propose a convex approximation, demonstrating its superiority over current state-of-the-art methodologies in literature. Furthermore, we establish connections between robust hypothesis testing and regularized formulations of non-robust risk functions, offering insightful interpretations. Our numerical study highlights the satisfactory testing performance and computational efficiency of the proposed framework.
comment: 26 pages, 2 figures
☆ CASPER: Carbon-Aware Scheduling and Provisioning for Distributed Web Services
There has been a significant societal push towards sustainable practices, including in computing. Modern interactive workloads such as geo-distributed web-services exhibit various spatiotemporal and performance flexibility, enabling the possibility to adapt the location, time, and intensity of processing to align with the availability of renewable and low-carbon energy. An example is a web application hosted across multiple cloud regions, each with varying carbon intensity based on their local electricity mix. Distributed load-balancing enables the exploitation of low-carbon energy through load migration across regions, reducing web applications carbon footprint. In this paper, we present CASPER, a carbon-aware scheduling and provisioning system that primarily minimizes the carbon footprint of distributed web services while also respecting their Service Level Objectives (SLO). We formulate CASPER as an multi-objective optimization problem that considers both the variable carbon intensity and latency constraints of the network. Our evaluation reveals the significant potential of CASPER in achieving substantial reductions in carbon emissions. Compared to baseline methods, CASPER demonstrates improvements of up to 70% with no latency performance degradation.
♻ ☆ Complexity of trust-region methods with unbounded Hessian approximations for smooth and nonsmooth optimization
We develop a worst-case evaluation complexity bound for trust-region methods in the presence of unbounded Hessian approximations. We use the algorithm of arXiv:2103.15993v3 as a model, which is designed for nonsmooth regularized problems, but applies to unconstrained smooth problems as a special case. Our analysis assumes that the growth of the Hessian approximation is controlled by the number of successful iterations. We show that the best known complexity bound of $\epsilon^{-2}$ deteriorates to $\epsilon^{-2/(1-p)}$, where $0 \le p < 1$ is a parameter that controls the growth of the Hessian approximation. The faster the Hessian approximation grows, the more the bound deteriorates. We construct an objective that satisfies all of our assumptions and for which our complexity bound is attained, which establishes that our bound is sharp. Numerical experiments conducted in double precision arithmetic are consistent with the analysis.
comment: 20 pages, 2 figures, 1 table
♻ ☆ Probabilistic Lookahead Strong Branching via a Stochastic Abstract Branching Model
Strong Branching (SB) is a cornerstone of all modern branching rules used in the Branch-and-Bound (BnB) algorithm, which is at the center of Mixed-Integer Programming solvers. In its full form, SB evaluates all variables to branch on and then selects the one producing the best relaxation, leading to small trees, but high runtimes. State-of-the-art branching rules therefore use SB with working limits to achieve both small enough trees and short run times. So far, these working limits have been established empirically. In this paper, we introduce a theoretical approach to guide how much SB to use at each node within the BnB. We first define an abstract stochastic tree model of the BnB algorithm where the geometric mean dual gains of all variables follow a given probability distribution. This model allows us to relate expected dual gains to tree sizes and explicitly compare the cost of sampling an additional SB candidate with the reward in expected tree size reduction. We then leverage the insight from the abstract model to design a new stopping criterion for SB, which fits a distribution to the dual gains and, at each node, dynamically continues or interrupts SB. This algorithm, which we refer to as Probabilistic Lookahead Strong Branching, improves both the tree size and runtime over MIPLIB instances, providing evidence that the method not only changes the amount of SB, but allocates it better.
♻ ☆ Sampling-Based Methods for Multi-Block Optimization Problems over Transport Polytopes
This paper focuses on multi-block optimization problems over transport polytopes, which underlie various applications including strongly correlated quantum physics and machine learning. Conventional block coordinate descent-type methods for the general multi-block problems store and operate on the matrix variables directly, resulting in formidable expenditure for large-scale settings. On the other hand, optimal transport problems, as a special case, have attracted extensive attention and numerical techniques that waive the use of the full matrices have recently emerged. However, it remains nontrivial to apply these techniques to the multi-block, possibly nonconvex problems with theoretical guarantees. In this work, we leverage the benefits of both sides and develop novel sampling-based block coordinate descent-type methods, which are equipped with either entropy regularization or Kullback-Leibler divergence. Each iteration of these methods solves subproblems restricted on the sampled degrees of freedom. Consequently, they involve only sparse matrices, which amounts to considerable complexity reductions. We explicitly characterize the sampling-induced errors and establish convergence and asymptotic properties for the methods equipped with the entropy regularization. Numerical experiments on typical strongly correlated electron systems corroborate their superior scalability over the methods utilizing full matrices. The advantage also enables the first visualization of approximate optimal transport maps between electron positions in three-dimensional contexts.
comment: 38 pages, 11 figures, 5 tables
♻ ☆ Data-Driven Output Prediction and Control of Stochastic Systems: An Innovation-Based Approach
Recent years have witnessed a booming interest in data-driven control of dynamical systems. However, the implicit data-driven output predictors are vulnerable to uncertainty such as process disturbance and measurement noise, causing unreliable predictions and unexpected control actions. In this brief, we put forward a new data-driven approach to output prediction of stochastic linear time-invariant (LTI) systems. By utilizing the innovation form, the uncertainty in stochastic LTI systems is recast as innovations that can be readily estimated from input-output data without knowing system matrices. In this way, by applying the fundamental lemma to the innovation form, we propose a new innovation-based data-driven output predictor (OP) of stochastic LTI systems, which bypasses the need for identifying state-space matrices explicitly and building a state estimator. The boundedness of the second moment of prediction errors in closed-loop is established under mild conditions. The proposed data-driven OP can be integrated into optimal control design for better performance. Numerical simulations are carried out to demonstrate the outperformance of the proposed innovation-based methods in output prediction and control design over existing formulations.
comment: Submitted to Automatica
♻ ☆ An iterative sample scenario approach for the dynamic dispatch waves problem
A challenge in same-day delivery operations is that delivery requests are typically not known beforehand, but are instead revealed dynamically during the day. This uncertainty introduces a trade-off between dispatching vehicles to serve requests as soon as they are revealed to ensure timely delivery, and delaying the dispatching decision to consolidate routing decisions with future, currently unknown requests. In this paper, we study the dynamic dispatch waves problem, a same-day delivery problem in which vehicles are dispatched at fixed decision moments. At each decision moment, the system operator must decide which of the known requests to dispatch, and how to route these dispatched requests. The operator's goal is to minimize the total routing cost while ensuring that all requests are served on time. We propose iterative conditional dispatch (ICD), an iterative solution construction procedure based on a sample scenario approach. ICD iteratively solves sample scenarios to classify requests to be dispatched, postponed, or undecided. The set of undecided requests shrinks in each iteration until a final dispatching decision is made in the last iteration. We develop two variants of ICD: one variant based on thresholds, and another variant based on similarity. A significant strength of ICD is that it is conceptually simple and easy to implement. This simplicity does not harm performance: through rigorous numerical experiments, we show that both variants efficiently navigate the large state and action spaces of the dynamic dispatch waves problem and quickly converge to a high-quality solution. Finally, we demonstrate that the threshold-based ICD variant achieves excellent results on instances from the EURO meets NeurIPS 2022 vehicle routing competition, nearly matching the performance of the winning machine learning-based strategy.
♻ ☆ Posterior Sampling Based on Gradient Flows of the MMD with Negative Distance Kernel ICLR 2024
We propose conditional flows of the maximum mean discrepancy (MMD) with the negative distance kernel for posterior sampling and conditional generative modeling. This MMD, which is also known as energy distance, has several advantageous properties like efficient computation via slicing and sorting. We approximate the joint distribution of the ground truth and the observations using discrete Wasserstein gradient flows and establish an error bound for the posterior distributions. Further, we prove that our particle flow is indeed a Wasserstein gradient flow of an appropriate functional. The power of our method is demonstrated by numerical examples including conditional image generation and inverse problems like superresolution, inpainting and computed tomography in low-dose and limited-angle settings.
comment: Published as a conference paper at ICLR 2024
♻ ☆ Neural Wasserstein Gradient Flows for Maximum Mean Discrepancies with Riesz Kernels ICML 2023
Wasserstein gradient flows of maximum mean discrepancy (MMD) functionals with non-smooth Riesz kernels show a rich structure as singular measures can become absolutely continuous ones and conversely. In this paper we contribute to the understanding of such flows. We propose to approximate the backward scheme of Jordan, Kinderlehrer and Otto for computing such Wasserstein gradient flows as well as a forward scheme for so-called Wasserstein steepest descent flows by neural networks (NNs). Since we cannot restrict ourselves to absolutely continuous measures, we have to deal with transport plans and velocity plans instead of usual transport maps and velocity fields. Indeed, we approximate the disintegration of both plans by generative NNs which are learned with respect to appropriate loss functions. In order to evaluate the quality of both neural schemes, we benchmark them on the interaction energy. Here we provide analytic formulas for Wasserstein schemes starting at a Dirac measure and show their convergence as the time step size tends to zero. Finally, we illustrate our neural MMD flows by numerical examples.
comment: Accepted at ICML 2023
♻ ☆ On continuation and convex Lyapunov functions
Suppose that the origin is globally asymptotically stable under a set of continuous vector fields on Euclidean space and suppose that all those vector fields come equipped with -- possibly different -- convex Lyapunov functions. We show that this implies there is a homotopy between any two of those vector fields such that the origin remains globally asymptotically stable along the homotopy. Relaxing the assumption on the origin to any compact convex set or relaxing convexity to geodesic convexity does not alter the conclusion. Imposing the same convexity assumptions on control Lyapunov functions leads to a Hautus-like stabilizability test. These results ought to be of interest in the context of learning stability certificates, policy gradient methods and switched systems.
comment: Final version, 12 pages, to appear in the IEEE Transactions on Automatic Control
♻ ☆ Input-output linearization and decoupling of mechanical control systems
In this work, we present a problem of simultaneous input-output feedback linearization and decoupling (non-interacting) for mechanical control systems with outputs. We show that the natural requirement of preserving mechanical structure of the system and of transformations imposes supplementary conditions when compared to the classical solution of the same problem for general control systems. These conditions can be expressed using objects on the configuration space only. We illustrate our results with several examples of mechanical control systems.
♻ ☆ An efficient sieving based secant method for sparse optimization problems with least-squares constraints
In this paper, we propose an efficient sieving based secant method to address the computational challenges of solving sparse optimization problems with least-squares constraints. A level-set method has been introduced in [X. Li, D.F. Sun, and K.-C. Toh, SIAM J. Optim., 28 (2018), pp. 1842--1866] that solves these problems by using the bisection method to find a root of a univariate nonsmooth equation $\varphi(\lambda) = \varrho$ for some $\varrho > 0$, where $\varphi(\cdot)$ is the value function computed by a solution of the corresponding regularized least-squares optimization problem. When the objective function in the constrained problem is a polyhedral gauge function, we prove that (i) for any positive integer $k$, $\varphi(\cdot)$ is piecewise $C^k$ in an open interval containing the solution $\lambda^*$ to the equation $\varphi(\lambda) = \varrho$; (ii) the Clarke Jacobian of $\varphi(\cdot)$ is always positive. These results allow us to establish the essential ingredients of the fast convergence rates of the secant method. Moreover, an adaptive sieving technique is incorporated into the secant method to effectively reduce the dimension of the level-set subproblems for computing the value of $\varphi(\cdot)$. The high efficiency of the proposed algorithm is demonstrated by extensive numerical results.
♻ ☆ On the Congruency-Constrained Matroid Base
Consider a matroid where all elements are labeled with an element in $\mathbb{Z}$. We are interested in finding a base where the sum of the labels is congruent to $g \pmod m$. We show that this problem can be solved in $\tilde{O}(2^{4m} n r^{5/6})$ time for a matroid with $n$ elements and rank $r$, when $m$ is either the product of two primes or a prime power. The algorithm can be generalized to all moduli and, in fact, to all abelian groups if a classic additive combinatorics conjecture by Schrijver and Seymour holds true. We also discuss the optimization version of the problem.
♻ ☆ Riemannian preconditioned coordinate descent for low multi-linear rank approximation
This paper presents a memory efficient, first-order method for low multi-linear rank approximation of high-order, high-dimensional tensors. In our method, we exploit the second-order information of the cost function and the constraints to suggest a new Riemannian metric on the Grassmann manifold. We use a Riemmanian coordinate descent method for solving the problem, and also provide a global convergence analysis matching that of the coordinate descent method in the Euclidean setting. We also show that each step of our method with the unit step-size is actually a step of the orthogonal iteration algorithm. Experimental results show the computational advantage of our method for high-dimensional tensors.
comment: 22 pages, 3 figures
♻ ☆ Callable convertible bonds under liquidity constraints and hybrid priorities
This paper investigates the callable convertible bond problem in the presence of a liquidity constraint modelled by Poisson signals. We assume that neither the bondholder nor the firm has absolute priority when they stop the game simultaneously, but instead, a proportion $m\in[0,1]$ of the bond is converted to the firm's stock and the rest is called by the firm. The paper thus generalizes the special case studied in [Liang and Sun, Dynkin games with Poisson random intervention times, SIAM Journal on Control and Optimization, 57 (2019), 2962-2991] where the bondholder has priority ($m=1$), and presents a complete solution to the callable convertible bond problem with liquidity constraint. The callable convertible bond is an example of a Dynkin game, but falls outside the standard paradigm since the payoffs do not depend in an ordered way upon which agent stops the game. We show how to deal with this non-ordered situation by introducing a new technique which may be of interest in its own right, and then apply it to the bond problem.
comment: 45 pages
♻ ☆ Don't Explain Noise: Robust Counterfactuals for Randomized Ensembles
Counterfactual explanations describe how to modify a feature vector in order to flip the outcome of a trained classifier. Obtaining robust counterfactual explanations is essential to provide valid algorithmic recourse and meaningful explanations. We study the robustness of explanations of randomized ensembles, which are always subject to algorithmic uncertainty even when the training data is fixed. We formalize the generation of robust counterfactual explanations as a probabilistic problem and show the link between the robustness of ensemble models and the robustness of base learners. We develop a practical method with good empirical performance and support it with theoretical guarantees for ensembles of convex base learners. Our results show that existing methods give surprisingly low robustness: the validity of naive counterfactuals is below $50\%$ on most data sets and can fall to $20\%$ on problems with many features. In contrast, our method achieves high robustness with only a small increase in the distance from counterfactual explanations to their initial observations.
♻ ☆ Stabilizing reinforcement learning control: A modular framework for optimizing over all stable behavior
We propose a framework for the design of feedback controllers that combines the optimization-driven and model-free advantages of deep reinforcement learning with the stability guarantees provided by using the Youla-Kucera parameterization to define the search domain. Recent advances in behavioral systems allow us to construct a data-driven internal model; this enables an alternative realization of the Youla-Kucera parameterization based entirely on input-output exploration data. Perhaps of independent interest, we formulate and analyze the stability of such data-driven models in the presence of noise. The Youla-Kucera approach requires a stable "parameter" for controller design. For the training of reinforcement learning agents, the set of all stable linear operators is given explicitly through a matrix factorization approach. Moreover, a nonlinear extension is given using a neural network to express a parameterized set of stable operators, which enables seamless integration with standard deep learning libraries. Finally, we show how these ideas can also be applied to tune fixed-structure controllers.
comment: Postprint; 31 pages. arXiv admin note: text overlap with arXiv:2304.03422
♻ ☆ Mean viability theorems and second-order Hamilton-Jacobi equations
We introduce the notion of mean viability for controlled stochastic differential equations and establish counterparts of Nagumo's classical viability theorems (necessary and sufficient conditions for mean viability). As an application, we provide a purely probabilistic proof of a comparison principle and of existence for contingent and viscosity solutions of second-order fully nonlinear path-dependent Hamilton-Jacobi-Bellman equations. We do not use compactness and optimal stopping arguments, which are usually employed in the literature on viscosity solutions for second-order path-dependent PDEs.
comment: 28 pages, to appear in SIAM Journal on Control and Optimization
♻ ☆ Gaussian Cooling and Dikin Walks: The Interior-Point Method for Logconcave Sampling
The connections between (convex) optimization and (logconcave) sampling have been considerably enriched in the past decade with many conceptual and mathematical analogies. For instance, the Langevin algorithm can be viewed as a sampling analogue of gradient descent and has condition-number-dependent guarantees on its performance. In the early 1990s, Nesterov and Nemirovski developed the Interior-Point Method (IPM) for convex optimization based on self-concordant barriers, providing efficient algorithms for structured convex optimization, often faster than the general method. This raises the following question: can we develop an analogous IPM for structured sampling problems? In 2012, Kannan and Narayanan proposed the Dikin walk for uniformly sampling polytopes, and an improved analysis was given in 2020 by Laddha-Lee-Vempala. The Dikin walk uses a local metric defined by a self-concordant barrier for linear constraints. Here we generalize this approach by developing and adapting IPM machinery together with the Dikin walk for poly-time sampling algorithms. Our IPM-based sampling framework provides an efficient warm start and goes beyond uniform distributions and linear constraints. We illustrate the approach on important special cases, in particular giving the fastest algorithms to sample uniform, exponential, or Gaussian distributions on a truncated PSD cone. The framework is general and can be applied to other sampling algorithms.
comment: Improved writing with minor errors fixed
♻ ☆ The stochastic Ravine accelerated gradient method with general extrapolation coefficients
In a real Hilbert space domain setting, we study the convergence properties of the stochastic Ravine accelerated gradient method for convex differentiable optimization. We consider the general form of this algorithm where the extrapolation coefficients can vary with each iteration, and where the evaluation of the gradient is subject to random errors. This general treatment models a breadth of practical algorithms and numerical implementations. We show that, under a proper tuning of the extrapolation parameters, and when the error variance associated with the gradient evaluations or the step-size sequences vanish sufficiently fast, the Ravine method provides fast convergence of the values both in expectation and almost surely. We also improve the convergence rates from O(.) to o(.). Moreover, we show almost sure summability property of the gradients, which implies the fast convergence of the gradients towards zero. This property reflects the fact that the high-resolution ODE of the Ravine method includes a Hessian-driven damping term. When the space is also separable, our analysis allows also to establish almost sure weak convergence of the sequence of iterates provided by the algorithm. We finally specialize the analysis to consider different parameter choices, including vanishing and constant (heavy ball method with friction) damping parameter, and present a comprehensive landscape of the tradeoffs in speed and accuracy associated with these parameter choices and statistical properties on the sequence of errors in the gradient computations. We provide a thorough discussion of the similarities and differences with the Nesterov accelerated gradient which satisfies similar asymptotic convergence rates.
comment: The insight and motivation for the study of the Ravine method and many of the derivations were the work of our beloved friend and colleague Hedy Attouch. As one of the final contributions of his long and illustrious career before his unfortunate recent departure, the other authors hope that the polished manuscript is an appreciated step in honoring his legacy
♻ ☆ Markov Decision Process Design: A Framework for Integrating Strategic and Operational Decisions
We consider the problem of optimally designing a system for repeated use under uncertainty. We develop a modeling framework that integrates design and operational phases, which are represented by a mixed-integer program and discounted-cost infinite-horizon Markov decision processes, respectively. We seek to simultaneously minimize the design costs and the subsequent expected operational costs. This problem setting arises naturally in several application areas, as we illustrate through examples. We derive a bilevel mixed-integer linear programming formulation for the problem and perform a computational study to demonstrate that realistic instances can be solved numerically.
♻ ☆ A Finitely Convergent Cutting Plane, and a Bender's Decomposition Algorithm for Mixed-Integer Convex and Two-Stage Convex Programs using Cutting Planes
We present a finitely convergent cutting-plane algorithm for solving a general mixed-integer convex programs given an oracle for solving general convex programs. This method is extended to solve a family of two-stage mixed-integer convex programs using cutting planes, with applications to solving distributionally-robust two-stage stochastic mixed-integer convex programs. Since algorithms purely using cutting planes are not very practical for implementation, we combined the cut generation with a branch-and-union scheme to develop a more practical algorithm. Analysis is also given for the case where convex programming oracle provides an $\epsilon-$optimal solution. Computational results on generated test problems show the practicality of our algorithm. Specifically, results show that addition of cuts speed up solution times by nearly 10-fold on the largest test problems that are solved.
comment: 50 pages, 2 tables and 1 figure
♻ ☆ On Some Mean Field Games and Master Equations through the lens of conservation laws
In this manuscript we derive a new nonlinear transport equation written on the space of probability measures that allows to study a class of deterministic mean field games and master equations, where the interaction of the agents happens only at the terminal time. The point of view via this transport equation has two important consequences. First, this equation reveals a new monotonicity condition that is sufficient both for the uniqueness of MFG Nash equilibria and for the global in time well-posedness of master equations. Interestingly, this condition is in general in dichotomy with both the Lasry--Lions and displacement monotonicity conditions, studied so far in the literature. Second, in the absence of monotonicity, the conservative form of the transport equation can be used to define weak entropy solutions to the master equation. We construct several concrete examples to demonstrate that MFG Nash equilibria, whether or not they actually exist, may not be selected by the entropy solutions of the master equation.
comment: final version; to appear in Math. Ann
♻ ☆ Mathematical Opportunities in Digital Twins (MATH-DT)
The report describes the discussions from the Workshop on Mathematical Opportunities in Digital Twins (MATH-DT) from December 11-13, 2023, George Mason University. It illustrates that foundational Mathematical advances are required for Digital Twins (DTs) that are different from traditional approaches. A traditional model, in biology, physics, engineering or medicine, starts with a generic physical law (e.g., equations) and is often a simplification of reality. A DT starts with a specific ecosystem, object or person (e.g., personalized care) representing reality, requiring multi -scale, -physics modeling and coupling. Thus, these processes begin at opposite ends of the simulation and modeling pipeline, requiring different reliability criteria and uncertainty assessments. Additionally, unlike existing approaches, a DT assists humans to make decisions for the physical system, which (via sensors) in turn feeds data into the DT, and operates for the life of the physical system. While some of the foundational mathematical research can be done without a specific application context, one must also keep specific applications in mind for DTs. E.g., modeling a bridge or a biological system (a patient), or a socio-technical system (a city) is very different. The models range from differential equations (deterministic/uncertain) in engineering, to stochastic in biology, including agent-based. These are multi-scale hybrid models or large scale (multi-objective) optimization problems under uncertainty. There are no universal models or approaches. For e.g., Kalman filters for forecasting might work in engineering, but can fail in biomedical domain. Ad hoc studies, with limited systematic work, have shown that AI/ML methods can fail for simple engineering systems and can work well for biomedical problems. A list of `Mathematical Opportunities and Challenges' concludes the report.
♻ ☆ Optimal First-Order Algorithms as a Function of Inequalities
In this work, we present a novel algorithm design methodology that finds the optimal algorithm as a function of inequalities. Specifically, we restrict convergence analyses of algorithms to use a prespecified subset of inequalities, rather than utilizing all true inequalities, and find the optimal algorithm subject to this restriction. This methodology allows us to design algorithms with certain desired characteristics. As concrete demonstrations of this methodology, we find new state-of-the-art accelerated first-order gradient methods using randomized coordinate updates and backtracking line searches.
Systems and Control 35
☆ SDP Synthesis of Maximum Coverage Trees for Probabilistic Planning under Control Constraints
The paper presents Maximal Covariance Backward Reachable Trees (MAXCOVAR BRT), which is a multi-query algorithm for planning of dynamic systems under stochastic motion uncertainty and constraints on the control input with explicit coverage guarantees. In contrast to existing roadmap-based probabilistic planning methods that sample belief nodes randomly and draw edges between them \cite{csbrm_tro2024}, under control constraints, the reachability of belief nodes needs to be explicitly established and is determined by checking the feasibility of a non-convex program. Moreover, there is no explicit consideration of coverage of the roadmap while adding nodes and edges during the construction procedure for the existing methods. Our contribution is a novel optimization formulation to add nodes and construct the corresponding edge controllers such that the generated roadmap results in provably maximal coverage under control constraints as compared to any other method of adding nodes and edges. We characterize formally the notion of coverage of a roadmap in this stochastic domain via introduction of the h-$\operatorname{BRS}$ (Backward Reachable Set of Distributions) of a tree of distributions under control constraints, and also support our method with extensive simulations on a 6 DoF model.
☆ Learning Hierarchical Control For Constrained Dynamic Task Assignment
This paper introduces a novel data-driven hierarchical control scheme for managing a fleet of nonlinear, capacity-constrained autonomous agents in an iterative environment. We propose a control framework consisting of a high-level dynamic task assignment and routing layer and low-level motion planning and tracking layer. Each layer of the control hierarchy uses a data-driven MPC policy, maintaining bounded computational complexity at each calculation of a new task assignment or actuation input. We utilize collected data to iteratively refine estimates of agent capacity usage, and update MPC policy parameters accordingly. Our approach leverages tools from iterative learning control to integrate learning at both levels of the hierarchy, and coordinates learning between levels in order to maintain closed-loop feasibility and performance improvement of the connected architecture.
☆ Learning Hierarchical Control Systems for Autonomous Systems with Energy Constraints
This paper focuses on the design of hierarchical control architectures for autonomous systems with energy constraints. We focus on systems where energy storage limitations and slow recharge rates drastically affect the way the autonomous systems are operated. Using examples from space robotics and public transportation, we motivate the need for formally designed learning hierarchical control systems. We propose a learning control architecture which incorporates learning mechanisms at various levels of the control hierarchy to improve performance and resource utilization. The proposed hierarchical control scheme relies on high-level energy-aware task planning and assignment, complemented by a low-level predictive control mechanism responsible for the autonomous execution of tasks, including motion control and energy management. Simulation examples show the benefits and the limitations of the proposed architecture when learning is used to obtain a more energy-efficient task allocation.
☆ Optimizing queues with deadlines under infrequent monitoring
In this paper, we aim to improve the percentage of packets meeting their deadline in discrete-time M/M/1 queues with infrequent monitoring. More specifically, we look into policies that only monitor the system (and subsequently take actions) after a packet arrival. We model the system as an MDP and provide the optimal policy for some special cases. Furthermore, we introduce a heuristic algorithm called "AB-n" for general deadlines. Finally, we provide numerical results demonstrating the desirable performance of "AB-n" policies.
comment: 11 pages, 6 figures
☆ Designing Robust Linear Output Feedback Controller based on CLF-CBF framework via Linear~Programming(LP-CLF-CBF)
We consider the problem of designing output feedback controllers that use measurements from a set of landmarks to navigate through a cell-decomposable environment using duality, Control Lyapunov and Barrier Functions (CLF, CBF), and Linear Programming. We propose two objectives for navigating in an environment, one to traverse the environment by making loops and one by converging to a stabilization point while smoothing the transition between consecutive cells. We test our algorithms in a simulation environment, evaluating the robustness of the approach to practical conditions, such as bearing-only measurements, and measurements acquired with a camera with a limited field of view.
comment: arXiv admin note: text overlap with arXiv:2203.04416
☆ Constrained Reinforcement Learning with Smoothed Log Barrier Function
Reinforcement Learning (RL) has been widely applied to many control tasks and substantially improved the performances compared to conventional control methods in many domains where the reward function is well defined. However, for many real-world problems, it is often more convenient to formulate optimization problems in terms of rewards and constraints simultaneously. Optimizing such constrained problems via reward shaping can be difficult as it requires tedious manual tuning of reward functions with several interacting terms. Recent formulations which include constraints mostly require a pre-training phase, which often needs human expertise to collect data or assumes having a sub-optimal policy readily available. We propose a new constrained RL method called CSAC-LB (Constrained Soft Actor-Critic with Log Barrier Function), which achieves competitive performance without any pre-training by applying a linear smoothed log barrier function to an additional safety critic. It implements an adaptive penalty for policy learning and alleviates the numerical issues that are known to complicate the application of the log barrier function method. As a result, we show that with CSAC-LB, we achieve state-of-the-art performance on several constrained control tasks with different levels of difficulty and evaluate our methods in a locomotion task on a real quadruped robot platform.
☆ Meta-learning of data-driven controllers with automatic model reference tuning: theory and experimental case study
Data-driven control offers a viable option for control scenarios where constructing a system model is expensive or time-consuming. Nonetheless, many of these algorithms are not entirely automated, often necessitating the adjustment of multiple hyperparameters through cumbersome trial-and-error processes and demanding significant amounts of data. In this paper, we explore a meta-learning approach to leverage potentially existing prior knowledge about analogous (though not identical) systems, aiming to reduce both the experimental workload and ease the tuning of the available degrees of freedom. We validate this methodology through an experimental case study involving the tuning of proportional, integral (PI) controllers for brushless DC (BLDC) motors with variable loads and architectures.
☆ Synthesizing Controller for Safe Navigation using Control Density Function
We consider the problem of navigating a nonlinear dynamical system from some initial set to some target set while avoiding collision with an unsafe set. We extend the concept of density function to control density function (CDF) for solving navigation problems with safety constraints. The occupancy-based interpretation of the measure associated with the density function is instrumental in imposing the safety constraints. The navigation problem with safety constraints is formulated as a quadratic program (QP) using CDF. The existing approach using the control barrier function (CBF) also formulates the navigation problem with safety constraints as QP. One of the main advantages of the proposed QP using CDF compared to QP formulated using CBF is that both the convergence/stability and safety can be combined and imposed using the CDF. Simulation results involving the Duffing oscillator and safe navigation of Dubin car models are provided to verify the main findings of the paper.
☆ On the continuity and smoothness of the value function in reinforcement learning and optimal control
The value function plays a crucial role as a measure for the cumulative future reward an agent receives in both reinforcement learning and optimal control. It is therefore of interest to study how similar the values of neighboring states are, i.e., to investigate the continuity of the value function. We do so by providing and verifying upper bounds on the value function's modulus of continuity. Additionally, we show that the value function is always H\"older continuous under relatively weak assumptions on the underlying system and that non-differentiable value functions can be made differentiable by slightly "disturbing" the system.
☆ Exploiting Over-The-Air Consensus for Collision Avoidance and Formation Control in Multi-Agent Systems
This paper introduces a distributed control method for multi-agent robotic systems employing Over the Air Consensus (OTA-Consensus). Designed for agents with decoupled single-integrator dynamics, this approach aims at efficient formation achievement and collision avoidance. As a distinctive feature, it leverages OTA's ability to exploit interference in wireless channels, a property traditionally considered a drawback, thus enhancing communication efficiency among robots. An analytical proof of asymptotic convergence is established for systems with time-varying communication topologies represented by sequences of strongly connected directed graphs. Comparative evaluations demonstrate significant efficiency improvements over current state-of-the-art methods, especially in scenarios with a large number of agents.
comment: Submitted to CDC 2024
☆ A new control-oriented METANET model to encompass service stations on highways
In this paper, we propose the METANET with service station (METANET-s) model, a second-order macroscopic traffic model that, compared to the classical METANET, incorporates the dynamics of service stations on highways. Specifically, we employ the (so-called) store-and-forward links to model the stop of vehicles and the possible queue forming in the process of merging back into the highway mainstream. We explore the capability of the METANET-s to capture well both traffic back propagation and capacity drops, which are typically caused by the presence of vehicles joining again the mainstream traffic from the service station. Therefore, capturing these effects is crucial to improving the model's predictive capabilities. Finally, we perform a comparative analysis with the Cell Transmission Model with service station (CTM-s), showcasing that the METANET-s describes the traffic evolution much better than its first-order counterpart.
comment: 7 pages, 4 figures, to be published in European Control Conference (ECC) 2024
☆ A Benchmark for the Application of Distributed Control Techniques to the Electricity Network of the European Economic Area
The European Economic Area Electricity Network Benchmark (EEA-ENB) is a multi-area power system representing the European network of transmission systems for electricity to facilitate the application of distributed control techniques. In the EEA-ENB we consider the Load Frequency Control (LFC) problem in the presence of renewable energy sources (RESs), and energy storage systems (ESSs). RESs are known to cause instability in power networks due to their inertia-less and intermittent characteristics, while ESSs are introduced as a resource to mitigate the problem. In the EEA-ENB, particular attention is dedicated to Distributed Model Predictive Control (DMPC), whose application is often limited to small and homogeneous test cases due to the lack of standardized large-scale scenarios for testing, and due to the large computation time required to obtain a centralized MPC action for performance comparison with DMPC strategies under consideration. The second problem is exacerbated when the scale of the system grows. To address these challenges and to provide a real-world-based and control-independent benchmark, the EEA-ENB has been developed. The benchmark includes a centralized MPC strategy providing performance and computation time metrics to compare distributed control within a repeatable and realistic simulation environment.
comment: 20 pages, 19 figures, 22 references, book chapter
☆ A Control Barrier Function Composition Approach for Multi-Agent Systems in Marine Applications
The agents within a multi-agent system (MAS) operating in marine environments often need to utilize task payloads and avoid collisions in coordination, necessitating adherence to a set of relative-pose constraints, which may include field-of-view, line-of-sight, collision-avoidance, and range constraints. A nominal controller designed for reference tracking may not guarantee the marine MAS stays safe w.r.t. these constraints. To modify the nominal input as one that enforces safety, we introduce a framework to systematically encode the relative-pose constraints as nonsmooth control barrier functions (NCBFs) and combine them as a single NCBF using Boolean composition, which enables a simplified verification process compared to using the NCBFs individually. While other relative-pose constraint functions have explicit derivatives, the challenging line-of-sight constraint is encoded with the minimum distance function between the line-of-sight set and other agents, whose derivative is not explicit. Hence, existing safe control design methods that consider composite NCBFs cannot be applied. To address this challenge, we propose a novel quadratic program formulation based on the dual of the minimum distance problem and develop a new theory to ensure the resulting control input guarantees constraint satisfaction. Lastly, we validate the effectiveness of our proposed framework on a simulated large-scale marine MAS and a real-world marine MAS comprising one Unmanned Surface Vehicle and two Unmanned Underwater Vehicles.
comment: 11 pages, 8 figures
☆ Transformation-Free Fixed-Structure Model Reduction for LPV Systems
In this paper, we propose a model reduction technique for linear parameter varying (LPV) systems based on available tools for fixed-structure controller synthesis. We start by transforming a model reduction problem into an equivalent controller synthesis problem by defining an appropriate generalized plant. The controller synthesis problem is then solved by using gradient-based tools available in the literature. Owing to the flexibility of the gradient-based synthesis tools, we are able to impose a desired structure on the obtained reduced model. Additionally, we obtain a bound on the approximation error as a direct output of the optimization problem. The proposed methods are applied on a benchmark mechanical system of interconnected masses, springs and dampers. To evaluate the effect of the proposed model-reduction approach on controller design, LPV controllers designed using the reduced models (with and without an imposed structure) are compared in closed-loop with the original model.
☆ Multi Methods of Matrix Analysis Use for Control and Optimization system in Control Engineering
Matrix analysis plays a crucial role in the field of control engineering, providing a powerful mathematical framework for the analysis and design of control systems. This research report explores various applications of matrix analysis in control engineering, focusing on its contributions to system modeling, stability analysis, controllablity, observability, and optimization. The report also discusses specific examples and case studies to illustrate the practical significance of matrix analysis in addressing real-world control engineering challenges Analyze controllability. Informally, a system is controllable if we can construct a set of inputs that will drive the system to any given state. Analyze observability. Informally, observability means that by controlling the inputs and watching the outputs of a system we can determine what the states were. Optimal Control is a control method that aims to find the optimal control input to achieve the best performance of the system under certain constraints. This performance index can be the system output, energy consumption, time, etc.
comment: 8 pages
☆ Event-triggered Boundary Control of Mixed-autonomy Traffic
Control problems of mixed-autonomy traffic system consisting of both Human-driven Vehicles (HV) and Autonomous Vehicles (AV) have gained increasing attention. This paper is focused on suppressing traffic oscillations of the mixed-autonomy traffic system using boundary control design. The mixed traffic dynamics are described by a 4 x 4 hyperbolic partial differential equations (PDE) which governs propagation of four properties in traffic including density of HV, density of AV, friction between two classes of vehicles from driving interactions, and averaged velocity. We propose event-triggered boundary control design since control signal of traffic light on ramp or varying speed limit cannot be updated in a continuous time fashion. We apply event-triggered mechanism for a PDE backstepping controller and obtain dynamic triggering condition. Lyapunov analysis is conducted to prove the exponential stability of the closed loop system with the event-triggered controller. Numerical simulation demonstrates how car-following spacing of AV affects event-triggering mechanism of control input in mixed-autonomy traffic.
☆ Conservative Linear Envelopes for High-Dimensional, Hamilton-Jacobi Reachability for Nonlinear Systems via the Hopf Formula
Hamilton-Jacobi reachability (HJR) analysis provides a value function that encodes (1) the set of states from which a nonlinear system with bounded control inputs can reach a goal (or avoid a failure set) despite any bounded disturbance, and (2) the corresponding optimal control policy to reach (or avoid). Though powerful, traditional methods for HJR rely on dynamic programming and suffer from exponential computation growth with respect to state dimension. The recently favored Hopf formula mitigates this ''curse of dimensionality'' by providing an efficient and space-parallelizable approach for solving the reachability problem. However, the Hopf formula can only be applied to linear time-varying systems. To overcome this limitation, we show that the error between a nonlinear system and a linear model can be transformed into an adversarial bounded artificial disturbance, making an envelope of the true value. One may then solve the dimension-robust Hopf formula for a linear game with this ''antagonistic error" to perform guaranteed conservative reachability analysis and control synthesis of nonlinear systems; this can be done for problem formulations in which no other HJR method is both computationally feasible and guaranteed. In addition, we offer several technical methods for reducing conservativeness in the analysis. We demonstrate the theory by solving the safe linear envelope in the controlled Van der Pol system, where the true reachable set may be observed, and by solving a 5 agent (15D) pursuit-evasion game with Dubins cars.
☆ Lane level joint control of off-ramp and main line speed guidance on expressway in rainy weather
In the upstream of the exit ramp of the expressway, the speed limit difference leads to a significant deceleration of the vehicle in the area adjacent to the off-ramp. The friction coefficient of the road surface decreases under rainy weather, and the above deceleration process can easily lead to sideslip and rollover of the vehicle. Dynamic speed guidance is an effective way to improve the status quo. Currently, there is an emerging trend to utilize I2V technology and high-precision map technology for lane level speed guidance control. This paper presents an optimized joint control strategy for main line-off-ramp speed guidance, which can adjust the guidance speed in real time according to the rainfall intensity. At the same time, this paper designs a progressive deceleration strategy, which works together with the speed guidance control to ensure the safe deceleration of vehicles. The simulation results show that the proposed control strategy outperforms the fixed speed limit control in terms of improving the total traveled time (TTT), total traveled distance (TTD) and standard deviation of speed (SD). Sensitivity analysis shows that the proposed control strategy can improve performance with the increase of the compliance rate of drivers. The speed guidance control method established in this paper can improve the vehicle operation efficiency in the off-ramp area of the expressway and reduce the speed difference of each vehicle in rainy weather, which guarantee the safety of expressway driving in the rainy day.
comment: 103rd TRB Conference
☆ LR-FHSS Transceiver for Direct-to-Satellite IoT Communications: Design, Implementation, and Verification
This paper proposes a long range-frequency hopping spread spectrum (LR-FHSS) transceiver design for the Direct-to-Satellite Internet of Things (DtS-IoT) communication system. The DtS-IoT system has recently attracted attention as a promising nonterrestrial network (NTN) solution to provide high-traffic and low-latency data transfer services to IoT devices in global coverage. In particular, this study provides guidelines for the overall DtS-IoT system architecture and design details that conform to the Long Range Wide-Area Network (LoRaWAN). Furthermore, we also detail various DtS-IoT use cases. Considering the multiple low-Earth orbit (LEO) satellites, we developed the LR-FHSS transceiver to improve system efficiency, which is the first attempt in real satellite communication systems using LR-FHSS. Moreover, as an extension of our previous work with perfect synchronization, we applied a robust synchronization scheme against the Doppler effect and co-channel interference (CCI) caused by LEO satellite channel environments, including signal detection for the simultaneous reception of numerous frequency hopping signals and an enhanced soft-output-Viterbi-algorithm (SOVA) for the header and payload receptions. Lastly, we present proof-of-concept implementation and testbeds using an application-specific integrated circuit (ASIC) chipset and a field-programmable gate array (FPGA) that verify the performance of the proposed LR-FHSS transceiver design of DtS-IoT communication systems. The laboratory test results reveal that the proposed LR-FHSS-based framework with the robust synchronization technique can provide wide coverage, seamless connectivity, and high throughput communication links for the realization of future sixth-generation (6G) networks.
comment: 17pages, 23 figures
☆ Carbon Footprint Reduction for Sustainable Data Centers in Real-Time
As machine learning workloads significantly increase energy consumption, sustainable data centers with low carbon emissions are becoming a top priority for governments and corporations worldwide. This requires a paradigm shift in optimizing power consumption in cooling and IT loads, shifting flexible loads based on the availability of renewable energy in the power grid, and leveraging battery storage from the uninterrupted power supply in data centers, using collaborative agents. The complex association between these optimization strategies and their dependencies on variable external factors like weather and the power grid carbon intensity makes this a hard problem. Currently, a real-time controller to optimize all these goals simultaneously in a dynamic real-world setting is lacking. We propose a Data Center Carbon Footprint Reduction (DC-CFR) multi-agent Reinforcement Learning (MARL) framework that optimizes data centers for the multiple objectives of carbon footprint reduction, energy consumption, and energy cost. The results show that the DC-CFR MARL agents effectively resolved the complex interdependencies in optimizing cooling, load shifting, and energy storage in real-time for various locations under real-world dynamic weather and grid carbon intensity conditions. DC-CFR significantly outperformed the industry standard ASHRAE controller with a considerable reduction in carbon emissions (14.5%), energy usage (14.4%), and energy cost (13.7%) when evaluated over one year across multiple geographical regions.
☆ PE-GPT: A Physics-Informed Interactive Large Language Model for Power Converter Modulation Design
This paper proposes PE-GPT, a custom-tailored large language model uniquely adapted for power converter modulation design. By harnessing in-context learning and specialized tiered physics-informed neural networks, PE-GPT guides users through text-based dialogues, recommending actionable modulation parameters. The effectiveness of PE-GPT is validated through a practical design case involving dual active bridge converters, supported by hardware experimentation. This research underscores the transformative potential of large language models in power converter modulation design, offering enhanced accessibility, explainability, and efficiency, thereby setting a new paradigm in the field.
☆ Joint Planning of Charging Stations and Power Systems for Heavy-Duty Drayage Trucks
As global concerns about climate change intensify, the transition towards zero-emission freight is becoming increasingly vital. Drayage is an important segment of the freight system, typically involving the transport of goods from seaports or intermodal terminals to nearby warehouses. This sector significantly contributes to not only greenhouse gas emissions, but also pollution in densely populated areas. This study presents a holistic optimization model designed for an efficient transition to zero-emission drayage, offering cost-effective strategies for the coordinated investment planning for power systems, charging infrastructure, and electric drayage trucks. The model is validated in the Greater Los Angeles area, where regulatory goals are among the most ambitious. Furthermore, the model's design allows for easy adaptation to other regions. By focusing on drayage trucks, this study also paves the way for future research into other freight categories, establishing a foundation for a more extensive exploration in this field.
comment: 34 pages, 10 figures
☆ Robust Model Based Reinforcement Learning Using $\mathcal{L}_1$ Adaptive Control
We introduce $\mathcal{L}_1$-MBRL, a control-theoretic augmentation scheme for Model-Based Reinforcement Learning (MBRL) algorithms. Unlike model-free approaches, MBRL algorithms learn a model of the transition function using data and use it to design a control input. Our approach generates a series of approximate control-affine models of the learned transition function according to the proposed switching law. Using the approximate model, control input produced by the underlying MBRL is perturbed by the $\mathcal{L}_1$ adaptive control, which is designed to enhance the robustness of the system against uncertainties. Importantly, this approach is agnostic to the choice of MBRL algorithm, enabling the use of the scheme with various MBRL algorithms. MBRL algorithms with $\mathcal{L}_1$ augmentation exhibit enhanced performance and sample efficiency across multiple MuJoCo environments, outperforming the original MBRL algorithms, both with and without system noise.
☆ Model order reduction of deep structured state-space models: A system-theoretic approach
With a specific emphasis on control design objectives, achieving accurate system modeling with limited complexity is crucial in parametric system identification. The recently introduced deep structured state-space models (SSM), which feature linear dynamical blocks as key constituent components, offer high predictive performance. However, the learned representations often suffer from excessively large model orders, which render them unsuitable for control design purposes. The current paper addresses this challenge by means of system-theoretic model order reduction techniques that target the linear dynamical blocks of SSMs. We introduce two regularization terms which can be incorporated into the training loss for improved model order reduction. In particular, we consider modal $\ell_1$ and Hankel nuclear norm regularization to promote sparsity, allowing one to retain only the relevant states without sacrificing accuracy. The presented regularizers lead to advantages in terms of parsimonious representations and faster inference resulting from the reduced order models. The effectiveness of the proposed methodology is demonstrated using real-world ground vibration data from an aircraft.
☆ Transmission Benefits and Cost Allocation under Ambiguity
Disputes over cost allocation can present a significant barrier to investment in shared infrastructure. While it may be desirable to allocate cost in a way that corresponds to expected benefits, investments in long-lived projects are made under conditions of substantial uncertainty. In the context of electricity transmission, uncertainty combined with the inherent complexity of power systems analysis prevents the calculation of an estimated distribution of benefits that is agreeable to all participants. To analyze aspects of the cost allocation problem, we construct a model for transmission and generation expansion planning under uncertainty, enabling the identification of transmission investments as well as the calculation of benefits for users of the network. Numerical tests confirm the potential for realized benefits at the participant level to differ significantly from ex ante estimates. Based on the model and numerical tests we discuss several issues, including 1) establishing a valid counterfactual against which to measure benefits, 2) allocating cost to new and incumbent generators vs. solely allocating to loads, 3) calculating benefits at the portfolio vs. the individual project level, 4) identifying losers in a surplus-enhancing transmission expansion, and 5) quantifying the divergence between cost allocation decisions made ex ante and benefits realized ex post.
comment: 32 pages, 7 figures, 7 tables
♻ ☆ Deep learning reduces sensor requirements for gust rejection on a small uncrewed aerial vehicle morphing wing
There is a growing need for uncrewed aerial vehicles (UAVs) to operate in cities. However, the uneven urban landscape and complex street systems cause large-scale wind gusts that challenge the safe and effective operation of UAVs. Current gust alleviation methods rely on traditional control surfaces and computationally expensive modeling to select a control action, leading to a slower response. Here, we used deep reinforcement learning to create an autonomous gust alleviation controller for a camber-morphing wing. This method reduced gust impact by 84%, directly from real-time, on-board pressure signals. Notably, we found that gust alleviation using signals from only three pressure taps was statistically indistinguishable from using six signals. This reduced-sensor fly-by-feel control opens the door to UAV missions in previously inoperable locations.
♻ ☆ Adversarial Attacks and Defenses in Automated Control Systems: A Comprehensive Benchmark
Integrating machine learning into Automated Control Systems (ACS) enhances decision-making in industrial process management. One of the limitations to the widespread adoption of these technologies in industry is the vulnerability of neural networks to adversarial attacks. This study explores the threats in deploying deep learning models for fault diagnosis in ACS using the Tennessee Eastman Process dataset. By evaluating three neural networks with different architectures, we subject them to six types of adversarial attacks and explore five different defense methods. Our results highlight the strong vulnerability of models to adversarial samples and the varying effectiveness of defense strategies. We also propose a novel protection approach by combining multiple defense methods and demonstrate it's efficacy. This research contributes several insights into securing machine learning within ACS, ensuring robust fault diagnosis in industrial processes.
♻ ☆ Bearing-Constrained Leader-Follower Formation of Single-Integrators with Disturbance Rejection: Adaptive Variable-Structure Approaches
This paper studies the problem of stabilizing a leader-follower formation specified by a set of bearing constraints and being disturbed by unknown uniformly bounded disturbance. A set of leaders are positioned at the desired positions, while each follower agent is modeled by a single integrator with disturbance of which the upper bound is unavailable for the control design. Adaptive variable-structure formation control laws using only displacements or bearing vectors are provided to stabilize the agents to a desired formation. Thanks to the adaptive mechanism, the control laws require neither the information of the bearing Laplacian nor the directions and the upper bounds of the disturbance. It is further proved that when the leaders are moving with the same bounded uniformly continuous velocity, the moving target formation can still be achieved under the proposed control laws. Simulation results are also given to support the stability analysis.
comment: 18 pages, 6 figures. This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Optimizing the Electrical Interface for Large-Scale Color-Center Quantum Processors
Quantum processors based on color centers in diamond are promising candidates for future large-scale quantum computers thanks to their flexible optical interface, (relatively) high operating temperature, and high-fidelity operation. Similar to other quantum-computing platforms, the electrical interface required to control and read out such qubits may limit both the performance of the whole system and its scalability. To address this challenge, this work analyzes the requirements of the electrical interface and investigates how to efficiently implement the electronic controller in a scalable architecture comprising a large number of identical unit cells. Among the different discussed functionalities, a specific focus is devoted to the generation of the static and dynamic magnetic fields driving the electron and nuclear spins, because of their major impact on fidelity and scalability. Following the derived requirements, different system architectures, such as a qubit frequency-multiplexing scheme, are considered to identify the most power efficient approach, especially in the presence of inhomogeneity of the qubit Larmor frequency across the processor. As a result, a non-frequency-multiplexed, 1-mm$^2$ unit-cell architecture is proposed as the optimal solution, able to address up to one electron-spin qubit and 9 nuclear-spin qubits within a 3-mW average power consumption, thus establishing the baseline for the scalable electrical interface for future large-scale color-center quantum computers.
comment: 20 pages, 10 figures; Revision 21-03-2024: Corrected affiliation, corrected typo Fig. 2, corrected DC noise PSD value (Z-field noise, X/Y-field noise) in Tab.1
♻ ☆ Input-output linearization and decoupling of mechanical control systems
In this work, we present a problem of simultaneous input-output feedback linearization and decoupling (non-interacting) for mechanical control systems with outputs. We show that the natural requirement of preserving mechanical structure of the system and of transformations imposes supplementary conditions when compared to the classical solution of the same problem for general control systems. These conditions can be expressed using objects on the configuration space only. We illustrate our results with several examples of mechanical control systems.
♻ ☆ Stabilizing reinforcement learning control: A modular framework for optimizing over all stable behavior
We propose a framework for the design of feedback controllers that combines the optimization-driven and model-free advantages of deep reinforcement learning with the stability guarantees provided by using the Youla-Kucera parameterization to define the search domain. Recent advances in behavioral systems allow us to construct a data-driven internal model; this enables an alternative realization of the Youla-Kucera parameterization based entirely on input-output exploration data. Perhaps of independent interest, we formulate and analyze the stability of such data-driven models in the presence of noise. The Youla-Kucera approach requires a stable "parameter" for controller design. For the training of reinforcement learning agents, the set of all stable linear operators is given explicitly through a matrix factorization approach. Moreover, a nonlinear extension is given using a neural network to express a parameterized set of stable operators, which enables seamless integration with standard deep learning libraries. Finally, we show how these ideas can also be applied to tune fixed-structure controllers.
comment: Postprint; 31 pages. arXiv admin note: text overlap with arXiv:2304.03422
♻ ☆ Representative Days and Hours with Piecewise Linear Transitions for Power System Planning
Electric demand and renewable power are highly variable, and the solution of a planning model relies on capturing this variability. This paper proposes a hybrid multi-area method that effectively captures both the intraday and interday chronology of real data considering extreme values, using a limited number of representative days, and time points within each day. An optimization-based representative extraction method is proposed to improve intraday chronology capturing. It ensures higher precision in preserving data chronology and extreme values than hierarchical clustering methods. The proposed method is based on a piecewise linear demand and supply representation, which reduces approximation errors compared to the traditional piecewise constant formulation. Additionally, sequentially linked day blocks with identical representatives, created through a mapping process, are employed for interday chronology capturing. To evaluate the efficiency of the proposed method, a comprehensive expansion co-planning model is developed, including transmission lines, energy storage systems, and wind farms.
♻ ☆ A Convex Parameterization of Controllers Constrained to use only Relative Measurements
The optimal controller design problem for systems equipped with sensors that measure only relative, rather than absolute, quantities is considered. This relative measurement structure is formulated as a design constraint; it is demonstrated that the resulting constrained controller design problem can be written as a convex program. Certain additional network structural constraints can be incorporated into this formulation, making it especially useful in distributed or networked settings. An illustrative example highlights the advantage of the proposed methodology over the standard formulation of the output feedback controller design problem. A numerical example is provided.
comment: 6 pages, 2 figures
♻ ☆ Markov Decision Process Design: A Framework for Integrating Strategic and Operational Decisions
We consider the problem of optimally designing a system for repeated use under uncertainty. We develop a modeling framework that integrates design and operational phases, which are represented by a mixed-integer program and discounted-cost infinite-horizon Markov decision processes, respectively. We seek to simultaneously minimize the design costs and the subsequent expected operational costs. This problem setting arises naturally in several application areas, as we illustrate through examples. We derive a bilevel mixed-integer linear programming formulation for the problem and perform a computational study to demonstrate that realistic instances can be solved numerically.
♻ ☆ Scalable Neural Dynamic Equivalence for Power Systems
Traditional grid analytics are model-based, relying strongly on accurate models of power systems, especially the dynamic models of generators, controllers, loads and other dynamic components. However, acquiring thorough power system models can be impractical in real operation due to inaccessible system parameters and privacy of consumers, which necessitate data-driven dynamic equivalencing of unknown subsystems. Learning reliable dynamic equivalent models for the external systems from SCADA and PMU data, however, is a long-standing intractable problem in power system analysis due to complicated nonlinearity and unforeseeable dynamic modes of power systems. This paper advances a practical application of neural dynamic equivalence (NeuDyE) called Driving Port NeuDyE (DP-NeuDyE), which exploits physics-informed machine learning and neural-ordinary-differential-equations (ODE-NET) to discover a dynamic equivalence of external power grids while preserving its dynamic behaviors after disturbances. The new contributions are threefold: A NeuDyE formulation to enable a continuous-time, data-driven dynamic equivalence of power systems, saving the effort and expense of acquiring inaccessible system; An introduction of a Physics-Informed NeuDyE learning (PI-NeuDyE) to actively control the closed-loop accuracy of NeuDyE; and A DP-NeuDyE to reduce the number of inputs required for the training. We conduct extensive case studies on the NPCC system to validate the generalizability and accuracy of both PI-NeuDyE and DP-NeuDyE, which span a multitude of scenarios, differing in the time required for fault clearance, the specific fault locations, and the limitations of data. Test results have demonstrated the scalability and practicality of NeuDyE, showing its potential to be used in ISO and utility control centers for online transient stability analysis and for planning purposes.
Robotics 71
☆ Natural Language as Polices: Reasoning for Coordinate-Level Embodied Control with LLMs
We demonstrate experimental results with LLMs that address robotics action planning problems. Recently, LLMs have been applied in robotics action planning, particularly using a code generation approach that converts complex high-level instructions into mid-level policy codes. In contrast, our approach acquires text descriptions of the task and scene objects, then formulates action planning through natural language reasoning, and outputs coordinate level control commands, thus reducing the necessity for intermediate representation code as policies. Our approach is evaluated on a multi-modal prompt simulation benchmark, demonstrating that our prompt engineering experiments with natural language reasoning significantly enhance success rates compared to its absence. Furthermore, our approach illustrates the potential for natural language descriptions to transfer robotics skills from known tasks to previously unseen tasks.
comment: 8 pages, 2 figures
☆ A Convex Formulation of Frictional Contact for the Material Point Method and Rigid Bodies
In this paper, we introduce a novel convex formulation that seamlessly integrates the Material Point Method (MPM) with articulated rigid body dynamics in frictional contact scenarios. We extend the linear corotational hyperelastic model into the realm of elastoplasticity and include an efficient return mapping algorithm. This approach is particularly effective for MPM simulations involving significant deformation and topology changes, while preserving the convexity of the optimization problem. Our method ensures global convergence, enabling the use of large simulation time steps without compromising robustness. We have validated our approach through rigorous testing and performance evaluations, highlighting its superior capabilities in managing complex simulations relevant to robotics. Compared to previous MPM based robotic simulators, our method significantly improves the stability of contact resolution -- a critical factor in robot manipulation tasks. We make our method available in the open-source robotics toolkit, Drake.
☆ Certified Human Trajectory Prediction
Trajectory prediction plays an essential role in autonomous vehicles. While numerous strategies have been developed to enhance the robustness of trajectory prediction models, these methods are predominantly heuristic and do not offer guaranteed robustness against adversarial attacks and noisy observations. In this work, we propose a certification approach tailored for the task of trajectory prediction. To this end, we address the inherent challenges associated with trajectory prediction, including unbounded outputs, and mutli-modality, resulting in a model that provides guaranteed robustness. Furthermore, we integrate a denoiser into our method to further improve the performance. Through comprehensive evaluations, we demonstrate the effectiveness of the proposed technique across various baselines and using standard trajectory prediction datasets. The code will be made available online: https://s-attack.github.io/
☆ Embedding Pose Graph, Enabling 3D Foundation Model Capabilities with a Compact Representation
This paper presents the Embedding Pose Graph (EPG), an innovative method that combines the strengths of foundation models with a simple 3D representation suitable for robotics applications. Addressing the need for efficient spatial understanding in robotics, EPG provides a compact yet powerful approach by attaching foundation model features to the nodes of a pose graph. Unlike traditional methods that rely on bulky data formats like voxel grids or point clouds, EPG is lightweight and scalable. It facilitates a range of robotic tasks, including open-vocabulary querying, disambiguation, image-based querying, language-directed navigation, and re-localization in 3D environments. We showcase the effectiveness of EPG in handling these tasks, demonstrating its capacity to improve how robots interact with and navigate through complex spaces. Through both qualitative and quantitative assessments, we illustrate EPG's strong performance and its ability to outperform existing methods in re-localization. Our work introduces a crucial step forward in enabling robots to efficiently understand and operate within large-scale 3D spaces.
☆ Projection-free computation of robust controllable sets with constrained zonotopes
We study the problem of computing robust controllable sets for discrete-time linear systems with additive uncertainty. We propose a tractable and scalable approach to inner- and outer-approximate robust controllable sets using constrained zonotopes, when the additive uncertainty set is a symmetric, convex, and compact set. Our least-squares-based approach uses novel closed-form approximations of the Pontryagin difference between a constrained zonotopic minuend and a symmetric, convex, and compact subtrahend. Unlike existing approaches, our approach does not rely on convex optimization solvers, and is projection-free for ellipsoidal and zonotopic uncertainty sets. We also propose a least-squares-based approach to compute a convex, polyhedral outer-approximation to constrained zonotopes, and characterize sufficient conditions under which all these approximations are exact. We demonstrate the computational efficiency and scalability of our approach in several case studies, including the design of abort-safe rendezvous trajectories for a spacecraft in near-rectilinear halo orbit under uncertainty. Our approach can inner-approximate a 20-step robust controllable set for a 100-dimensional linear system in under 15 seconds on a standard computer.
comment: 22 pages, 6 figures
Reinforcement Learning for Online Testing of Autonomous Driving Systems: a Replication and Extension Study
In a recent study, Reinforcement Learning (RL) used in combination with many-objective search, has been shown to outperform alternative techniques (random search and many-objective search) for online testing of Deep Neural Network-enabled systems. The empirical evaluation of these techniques was conducted on a state-of-the-art Autonomous Driving System (ADS). This work is a replication and extension of that empirical study. Our replication shows that RL does not outperform pure random test generation in a comparison conducted under the same settings of the original study, but with no confounding factor coming from the way collisions are measured. Our extension aims at eliminating some of the possible reasons for the poor performance of RL observed in our replication: (1) the presence of reward components providing contrasting or useless feedback to the RL agent; (2) the usage of an RL algorithm (Q-learning) which requires discretization of an intrinsically continuous state space. Results show that our new RL agent is able to converge to an effective policy that outperforms random testing. Results also highlight other possible improvements, which open to further investigations on how to best leverage RL for online ADS testing.
☆ DBA-Fusion: Tightly Integrating Deep Dense Visual Bundle Adjustment with Multiple Sensors for Large-Scale Localization and Mapping
Visual simultaneous localization and mapping (VSLAM) has broad applications, with state-of-the-art methods leveraging deep neural networks for better robustness and applicability. However, there is a lack of research in fusing these learning-based methods with multi-sensor information, which could be indispensable to push related applications to large-scale and complex scenarios. In this paper, we tightly integrate the trainable deep dense bundle adjustment (DBA) with multi-sensor information through a factor graph. In the framework, recurrent optical flow and DBA are performed among sequential images. The Hessian information derived from DBA is fed into a generic factor graph for multi-sensor fusion, which employs a sliding window and supports probabilistic marginalization. A pipeline for visual-inertial integration is firstly developed, which provides the minimum ability of metric-scale localization and mapping. Furthermore, other sensors (e.g., global navigation satellite system) are integrated for driftless and geo-referencing functionality. Extensive tests are conducted on both public datasets and self-collected datasets. The results validate the superior localization performance of our approach, which enables real-time dense mapping in large-scale environments. The code has been made open-source (https://github.com/GREAT-WHU/DBA-Fusion).
☆ What Matters for Active Texture Recognition With Vision-Based Tactile Sensors ICRA
This paper explores active sensing strategies that employ vision-based tactile sensors for robotic perception and classification of fabric textures. We formalize the active sampling problem in the context of tactile fabric recognition and provide an implementation of information-theoretic exploration strategies based on minimizing predictive entropy and variance of probabilistic models. Through ablation studies and human experiments, we investigate which components are crucial for quick and reliable texture recognition. Along with the active sampling strategies, we evaluate neural network architectures, representations of uncertainty, influence of data augmentation, and dataset variability. By evaluating our method on a previously published Active Clothing Perception Dataset and on a real robotic system, we establish that the choice of the active exploration strategy has only a minor influence on the recognition accuracy, whereas data augmentation and dropout rate play a significantly larger role. In a comparison study, while humans achieve 66.9% recognition accuracy, our best approach reaches 90.0% in under 5 touches, highlighting that vision-based tactile sensors are highly effective for fabric texture recognition.
comment: 7 pages, 9 figures, accepted at 2024 IEEE International Conference on Robotics and Automation (ICRA)
☆ Loss Regularizing Robotic Terrain Classification
Locomotion mechanics of legged robots are suitable when pacing through difficult terrains. Recognising terrains for such robots are important to fully yoke the versatility of their movements. Consequently, robotic terrain classification becomes significant to classify terrains in real time with high accuracy. The conventional classifiers suffer from overfitting problem, low accuracy problem, high variance problem, and not suitable for live dataset. On the other hand, classifying a growing dataset is difficult for convolution based terrain classification. Supervised recurrent models are also not practical for this classification. Further, the existing recurrent architectures are still evolving to improve accuracy of terrain classification based on live variable-length sensory data collected from legged robots. This paper proposes a new semi-supervised method for terrain classification of legged robots, avoiding preprocessing of long variable-length dataset. The proposed method has a stacked Long Short-Term Memory architecture, including a new loss regularization. The proposed method solves the existing problems and improves accuracy. Comparison with the existing architectures show the improvements.
comment: Preliminary draft of the work published in IEEE conference 2023
☆ DVMNet: Computing Relative Pose for Unseen Objects Beyond Hypotheses CVPR 2024
Determining the relative pose of an object between two images is pivotal to the success of generalizable object pose estimation. Existing approaches typically approximate the continuous pose representation with a large number of discrete pose hypotheses, which incurs a computationally expensive process of scoring each hypothesis at test time. By contrast, we present a Deep Voxel Matching Network (DVMNet) that eliminates the need for pose hypotheses and computes the relative object pose in a single pass. To this end, we map the two input RGB images, reference and query, to their respective voxelized 3D representations. We then pass the resulting voxels through a pose estimation module, where the voxels are aligned and the pose is computed in an end-to-end fashion by solving a least-squares problem. To enhance robustness, we introduce a weighted closest voxel algorithm capable of mitigating the impact of noisy voxels. We conduct extensive experiments on the CO3D, LINEMOD, and Objaverse datasets, demonstrating that our method delivers more accurate relative pose estimates for novel objects at a lower computational cost compared to state-of-the-art methods. Our code is released at: https://github.com/sailor-z/DVMNet/.
comment: Accepted by CVPR 2024
☆ Reward-Driven Automated Curriculum Learning for Interaction-Aware Self-Driving at Unsignalized Intersections
In this work, we present a reward-driven automated curriculum reinforcement learning approach for interaction-aware self-driving at unsignalized intersections, taking into account the uncertainties associated with surrounding vehicles (SVs). These uncertainties encompass the uncertainty of SVs' driving intention and also the quantity of SVs. To deal with this problem, the curriculum set is specifically designed to accommodate a progressively increasing number of SVs. By implementing an automated curriculum selection mechanism, the importance weights are rationally allocated across various curricula, thereby facilitating improved sample efficiency and training outcomes. Furthermore, the reward function is meticulously designed to guide the agent towards effective policy exploration. Thus the proposed framework could proactively address the above uncertainties at unsignalized intersections by employing the automated curriculum learning technique that progressively increases task difficulty, and this ensures safe self-driving through effective interaction with SVs. Comparative experiments are conducted in $Highway\_Env$, and the results indicate that our approach achieves the highest task success rate, attains strong robustness to initialization parameters of the curriculum selection module, and exhibits superior adaptability to diverse situational configurations at unsignalized intersections. Furthermore, the effectiveness of the proposed method is validated using the high-fidelity CARLA simulator.
comment: 8 pages, 6 figures
☆ LaCE-LHMP: Airflow Modelling-Inspired Long-Term Human Motion Prediction By Enhancing Laminar Characteristics in Human Flow ICRA
Long-term human motion prediction (LHMP) is essential for safely operating autonomous robots and vehicles in populated environments. It is fundamental for various applications, including motion planning, tracking, human-robot interaction and safety monitoring. However, accurate prediction of human trajectories is challenging due to complex factors, including, for example, social norms and environmental conditions. The influence of such factors can be captured through Maps of Dynamics (MoDs), which encode spatial motion patterns learned from (possibly scattered and partial) past observations of motion in the environment and which can be used for data-efficient, interpretable motion prediction (MoD-LHMP). To address the limitations of prior work, especially regarding accuracy and sensitivity to anomalies in long-term prediction, we propose the Laminar Component Enhanced LHMP approach (LaCE-LHMP). Our approach is inspired by data-driven airflow modelling, which estimates laminar and turbulent flow components and uses predominantly the laminar components to make flow predictions. Based on the hypothesis that human trajectory patterns also manifest laminar flow (that represents predictable motion) and turbulent flow components (that reflect more unpredictable and arbitrary motion), LaCE-LHMP extracts the laminar patterns in human dynamics and uses them for human motion prediction. We demonstrate the superior prediction performance of LaCE-LHMP through benchmark comparisons with state-of-the-art LHMP methods, offering an unconventional perspective and a more intuitive understanding of human movement patterns.
comment: Accepted to the 2024 IEEE International Conference on Robotics and Automation (ICRA)
☆ From One to Many: How Active Robot Swarm Sizes Influence Human Cognitive Processes
In robotics, understanding human interaction with autonomous systems is crucial for enhancing collaborative technologies. We focus on human-swarm interaction (HSI), exploring how differently sized groups of active robots affect operators' cognitive and perceptual reactions over different durations. We analyze the impact of different numbers of active robots within a 15-robot swarm on operators' time perception, emotional state, flow experience, and task difficulty perception. Our findings indicate that managing multiple active robots when compared to one active robot significantly alters time perception and flow experience, leading to a faster passage of time and increased flow. More active robots and extended durations cause increased emotional arousal and perceived task difficulty, highlighting the interaction between robot the number of active robots and human cognitive processes. These insights inform the creation of intuitive human-swarm interfaces and aid in developing swarm robotic systems aligned with human cognitive structures, enhancing human-robot collaboration.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
☆ Motion Generation from Fine-grained Textual Descriptions
The task of text2motion is to generate motion sequences from given textual descriptions, where a model should explore the interactions between natural language instructions and human body movements. While most existing works are confined to coarse-grained motion descriptions (e.g., "A man squats."), fine-grained ones specifying movements of relevant body parts are barely explored. Models trained with coarse texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure in generating motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset with fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with delicate prompts. Accordingly, we design a new text2motion model, FineMotionDiffuse, which makes full use of fine-grained textual information. Our experiments show that FineMotionDiffuse trained on FineHumanML3D acquires good results in quantitative evaluation. We also find this model can better generate spatially/chronologically composite motions by learning the implicit mappings from simple descriptions to the corresponding basic motions.
☆ Iterative Active-Inactive Obstacle Classification for Time-Optimal Collision Avoidance IROS24
Time-optimal obstacle avoidance is a prevalent problem encountered in various fields, including robotics and autonomous vehicles, where the task involves determining a path for a moving vehicle to reach its goal while navigating around obstacles within its environment. This problem becomes increasingly challenging as the number of obstacles in the environment rises. We propose an iterative active-inactive obstacle approach, which involves identifying a subset of the obstacles as "active", that considers solely the effect of the "active" obstacles on the path of the moving vehicle. The remaining obstacles are considered "inactive" and are not considered in the path planning process. The obstacles are classified as 'active' on the basis of previous findings derived from prior iterations. This approach allows for a more efficient calculation of the optimal path by reducing the number of obstacles that need to be considered. The effectiveness of the proposed method is demonstrated with two different dynamic models using the various number of obstacles. The results show that the proposed method is able to find the optimal path in a timely manner, while also being able to handle a large number of obstacles in the environment and the constraints on the motion of the object.
comment: This paper is under review in IROS24
☆ CLIPSwarm: Generating Drone Shows from Text Prompts with Vision-Language Models
This paper introduces CLIPSwarm, a new algorithm designed to automate the modeling of swarm drone formations based on natural language. The algorithm begins by enriching a provided word, to compose a text prompt that serves as input to an iterative approach to find the formation that best matches the provided word. The algorithm iteratively refines formations of robots to align with the textual description, employing different steps for "exploration" and "exploitation". Our framework is currently evaluated on simple formation targets, limited to contour shapes. A formation is visually represented through alpha-shape contours and the most representative color is automatically found for the input word. To measure the similarity between the description and the visual representation of the formation, we use CLIP [1], encoding text and images into vectors and assessing their similarity. Subsequently, the algorithm rearranges the formation to visually represent the word more effectively, within the given constraints of available drones. Control actions are then assigned to the drones, ensuring robotic behavior and collision-free movement. Experimental results demonstrate the system's efficacy in accurately modeling robot formations from natural language descriptions. The algorithm's versatility is showcased through the execution of drone shows in photorealistic simulation with varying shapes. We refer the reader to the supplementary video for a visual reference of the results.
☆ FACT: Fast and Active Coordinate Initialization for Vision-based Drone Swarms
Swarm robots have sparked remarkable developments across a range of fields. While it is necessary for various applications in swarm robots, a fast and robust coordinate initialization in vision-based drone swarms remains elusive. To this end, our paper proposes a complete system to recover a swarm's initial relative pose on platforms with size, weight, and power (SWaP) constraints. To overcome limited coverage of field-of-view (FoV), the drones rotate in place to obtain observations. To tackle the anonymous measurements, we formulate a non-convex rotation estimation problem and transform it into a semi-definite programming (SDP) problem, which can steadily obtain global optimal values. Then we utilize the Hungarian algorithm to recover relative translation and correspondences between observations and drone identities. To safely acquire complete observations, we actively search for positions and generate feasible trajectories to avoid collisions. To validate the practicability of our system, we conduct experiments on a vision-based drone swarm with only stereo cameras and inertial measurement units (IMUs) as sensors. The results demonstrate that the system can robustly get accurate relative poses in real time with limited onboard computation resources. The source code is released.
☆ Mobile Robot Localization: a Modular, Odometry-Improving Approach
Despite the number of works published in recent years, vehicle localization remains an open, challenging problem. While map-based localization and SLAM algorithms are getting better and better, they remain a single point of failure in typical localization pipelines. This paper proposes a modular localization architecture that fuses sensor measurements with the outputs of off-the-shelf localization algorithms. The fusion filter estimates model uncertainties to improve odometry in case absolute pose measurements are lost entirely. The architecture is validated experimentally on a real robot navigating autonomously proving a reduction of the position error of more than 90% with respect to the odometrical estimate without uncertainty estimation in a two-minute navigation period without position measurements.
comment: Accepted at IEEE European Control Conference 2024
☆ Fast-Poly: A Fast Polyhedral Framework For 3D Multi-Object Tracking
3D Multi-Object Tracking (MOT) captures stable and comprehensive motion states of surrounding obstacles, essential for robotic perception. However, current 3D trackers face issues with accuracy and latency consistency. In this paper, we propose Fast-Poly, a fast and effective filter-based method for 3D MOT. Building upon our previous work Poly-MOT, Fast-Poly addresses object rotational anisotropy in 3D space, enhances local computation densification, and leverages parallelization technique, improving inference speed and precision. Fast-Poly is extensively tested on two large-scale tracking benchmarks with Python implementation. On the nuScenes dataset, Fast-Poly achieves new state-of-the-art performance with 75.8% AMOTA among all methods and can run at 34.2 FPS on a personal CPU. On the Waymo dataset, Fast-Poly exhibits competitive accuracy with 63.6% MOTA and impressive inference speed (35.5 FPS). The source code is publicly available at https://github.com/lixiaoyu2000/FastPoly.
comment: 1st on the NuScenes Tracking benchmark with 75.8 AMOTA and 34.2 FPS
☆ Automatic Navigation Map Generation for Mobile Robots in Urban Environments
A fundamental prerequisite for safe and efficient navigation of mobile robots is the availability of reliable navigation maps upon which trajectories can be planned. With the increasing industrial interest in mobile robotics, especially in urban environments, the process of generating navigation maps has become of particular interest, being a labor intensive step of the deployment process. Automating this step is challenging and becomes even more arduous when the perception capabilities are limited by cost considerations. This paper proposes an algorithm to automatically generate navigation maps using a typical navigation-oriented sensor setup: a single top-mounted 3D LiDAR sensor. The proposed method is designed and validated with the urban environment as the main use case: it is shown to be able to produce accurate maps featuring different terrain types, positive obstacles of different heights as well as negative obstacles. The algorithm is applied to data collected in a typical urban environment with a wheeled inverted pendulum robot, showing its robustness against localization, perception and dynamic uncertainties. The generated map is validated against a human-made map.
☆ Caching-Augmented Lifelong Multi-Agent Path Finding
Multi-Agent Path Finding (MAPF), which involves finding collision-free paths for multiple robots, is crucial in various applications. Lifelong MAPF, where targets are reassigned to agents as soon as they complete their initial objectives, offers a more accurate approximation of real-world warehouse planning. In this paper, we present a novel mechanism named Caching-Augmented Lifelong MAPF (CAL-MAPF), designed to improve the performance of Lifelong MAPF. We have developed a new map grid type called cache for temporary item storage and replacement and designed a lock mechanism for it to improve the stability of the planning solution. This cache mechanism was evaluated using various cache replacement policies and a spectrum of input task distributions. We identified three main factors significantly impacting CAL-MAPF performance through experimentation: suitable input task distribution, high cache hit rate, and smooth traffic. Overall, CAL-MAPF has demonstrated potential for performance improvements in certain task distributions, maps and agent configurations.
☆ Unifying Local and Global Multimodal Features for Place Recognition in Aliased and Low-Texture Environments ICRA
Perceptual aliasing and weak textures pose significant challenges to the task of place recognition, hindering the performance of Simultaneous Localization and Mapping (SLAM) systems. This paper presents a novel model, called UMF (standing for Unifying Local and Global Multimodal Features) that 1) leverages multi-modality by cross-attention blocks between vision and LiDAR features, and 2) includes a re-ranking stage that re-orders based on local feature matching the top-k candidates retrieved using a global representation. Our experiments, particularly on sequences captured on a planetary-analogous environment, show that UMF outperforms significantly previous baselines in those challenging aliased environments. Since our work aims to enhance the reliability of SLAM in all situations, we also explore its performance on the widely used RobotCar dataset, for broader applicability. Code and models are available at https://github.com/DLR-RM/UMF
comment: Accepted submission to International Conference on Robotics and Automation (ICRA), 2024
☆ Centroidal State Estimation based on the Koopman Embedding for Dynamic Legged Locomotion IROS 2024
In this paper, we introduce a novel approach to centroidal state estimation, which plays a crucial role in predictive model-based control strategies for dynamic legged locomotion. Our approach uses the Koopman operator theory to transform the robot's complex nonlinear dynamics into a linear system, by employing dynamic mode decomposition and deep learning for model construction. We evaluate both models on their linearization accuracy and capability to capture both fast and slow dynamic system responses. We then select the most suitable model for estimation purposes, and integrate it within a moving horizon estimator. This estimator is formulated as a convex quadratic program, to facilitate robust, real-time centroidal state estimation. Through extensive simulation experiments on a quadruped robot executing various dynamic gaits, our data-driven framework outperforms conventional filtering techniques based on nonlinear dynamics. Our estimator addresses challenges posed by force/torque measurement noise in highly dynamic motions and accurately recovers the centroidal states, demonstrating the adaptability and effectiveness of the Koopman-based linear representation for complex locomotive behaviors. Importantly, our model based on dynamic mode decomposition, trained with two locomotion patterns (trot and jump), successfully estimates the centroidal states for a different motion (bound) without retraining.
comment: Submitted to IROS 2024
☆ ManiPose: A Comprehensive Benchmark for Pose-aware Object Manipulation in Robotics IROS 2024
Robotic manipulation in everyday scenarios, especially in unstructured environments, requires skills in pose-aware object manipulation (POM), which adapts robots' grasping and handling according to an object's 6D pose. Recognizing an object's position and orientation is crucial for effective manipulation. For example, if a mug is lying on its side, it's more effective to grasp it by the rim rather than the handle. Despite its importance, research in POM skills remains limited, because learning manipulation skills requires pose-varying simulation environments and datasets. This paper introduces ManiPose, a pioneering benchmark designed to advance the study of pose-varying manipulation tasks. ManiPose encompasses: 1) Simulation environments for POM feature tasks ranging from 6D pose-specific pick-and-place of single objects to cluttered scenes, further including interactions with articulated objects. 2) A comprehensive dataset featuring geometrically consistent and manipulation-oriented 6D pose labels for 2936 real-world scanned rigid objects and 100 articulated objects across 59 categories. 3) A baseline for POM, leveraging the inferencing abilities of LLM (e.g., ChatGPT) to analyze the relationship between 6D pose and task-specific requirements, offers enhanced pose-aware grasp prediction and motion planning capabilities. Our benchmark demonstrates notable advancements in pose estimation, pose-aware manipulation, and real-robot skill transfer, setting new standards for POM research. We will open-source the ManiPose benchmark with the final version paper, inviting the community to engage with our resources, available at our website:https://sites.google.com/view/manipose.
comment: 8 pages, 7 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)
☆ GeRM: A Generalist Robotic Model with Mixture-of-experts for Quadruped Robot
Multi-task robot learning holds significant importance in tackling diverse and complex scenarios. However, current approaches are hindered by performance issues and difficulties in collecting training datasets. In this paper, we propose GeRM (Generalist Robotic Model). We utilize offline reinforcement learning to optimize data utilization strategies to learn from both demonstrations and sub-optimal data, thus surpassing the limitations of human demonstrations. Thereafter, we employ a transformer-based VLA network to process multi-modal inputs and output actions. By introducing the Mixture-of-Experts structure, GeRM allows faster inference speed with higher whole model capacity, and thus resolves the issue of limited RL parameters, enhancing model performance in multi-task learning while controlling computational costs. Through a series of experiments, we demonstrate that GeRM outperforms other methods across all tasks, while also validating its efficiency in both training and inference processes. Additionally, we uncover its potential to acquire emergent skills. Additionally, we contribute the QUARD-Auto dataset, collected automatically to support our training approach and foster advancements in multi-task quadruped robot learning. This work presents a new paradigm for reducing the cost of collecting robot data and driving progress in the multi-task learning community.
☆ MULAN-WC: Multi-Robot Localization Uncertainty-aware Active NeRF with Wireless Coordination
This paper presents MULAN-WC, a novel multi-robot 3D reconstruction framework that leverages wireless signal-based coordination between robots and Neural Radiance Fields (NeRF). Our approach addresses key challenges in multi-robot 3D reconstruction, including inter-robot pose estimation, localization uncertainty quantification, and active best-next-view selection. We introduce a method for using wireless Angle-of-Arrival (AoA) and ranging measurements to estimate relative poses between robots, as well as quantifying and incorporating the uncertainty embedded in the wireless localization of these pose estimates into the NeRF training loss to mitigate the impact of inaccurate camera poses. Furthermore, we propose an active view selection approach that accounts for robot pose uncertainty when determining the next-best views to improve the 3D reconstruction, enabling faster convergence through intelligent view selection. Extensive experiments on both synthetic and real-world datasets demonstrate the effectiveness of our framework in theory and in practice. Leveraging wireless coordination and localization uncertainty-aware training, MULAN-WC can achieve high-quality 3d reconstruction which is close to applying the ground truth camera poses. Furthermore, the quantification of the information gain from a novel view enables consistent rendering quality improvement with incrementally captured images by commending the robot the novel view position. Our hardware experiments showcase the practicality of deploying MULAN-WC to real robotic systems.
☆ Discretizing SO(2)-Equivariant Features for Robotic Kitting
Robotic kitting has attracted considerable attention in logistics and industrial settings. However, existing kitting methods encounter challenges such as low precision and poor efficiency, limiting their widespread applications. To address these issues, we present a novel kitting framework that improves both the precision and computational efficiency of complex kitting tasks. Firstly, our approach introduces a fine-grained orientation estimation technique in the picking module, significantly enhancing orientation precision while effectively decoupling computational load from orientation granularity. This approach combines an SO(2)-equivariant network with a group discretization operation to preciously predict discrete orientation distributions. Secondly, we develop the Hand-tool Kitting Dataset (HKD) to evaluate the performance of different solutions in handling orientation-sensitive kitting tasks. This dataset comprises a diverse collection of hand tools and synthetically created kits, which reflects the complexities encountered in real-world kitting scenarios. Finally, a series of experiments are conducted to evaluate the performance of the proposed method. The results demonstrate that our approach offers remarkable precision and enhanced computational efficiency in robotic kitting tasks.
comment: 8 pages, 6 figures
☆ AMP: Autoregressive Motion Prediction Revisited with Next Token Prediction for Autonomous Driving
As an essential task in autonomous driving (AD), motion prediction aims to predict the future states of surround objects for navigation. One natural solution is to estimate the position of other agents in a step-by-step manner where each predicted time-step is conditioned on both observed time-steps and previously predicted time-steps, i.e., autoregressive prediction. Pioneering works like SocialLSTM and MFP design their decoders based on this intuition. However, almost all state-of-the-art works assume that all predicted time-steps are independent conditioned on observed time-steps, where they use a single linear layer to generate positions of all time-steps simultaneously. They dominate most motion prediction leaderboards due to the simplicity of training MLPs compared to autoregressive networks. In this paper, we introduce the GPT style next token prediction into motion forecasting. In this way, the input and output could be represented in a unified space and thus the autoregressive prediction becomes more feasible. However, different from language data which is composed of homogeneous units -words, the elements in the driving scene could have complex spatial-temporal and semantic relations. To this end, we propose to adopt three factorized attention modules with different neighbors for information aggregation and different position encoding styles to capture their relations, e.g., encoding the transformation between coordinate systems for spatial relativity while adopting RoPE for temporal relativity. Empirically, by equipping with the aforementioned tailored designs, the proposed method achieves state-of-the-art performance in the Waymo Open Motion and Waymo Interaction datasets. Notably, AMP outperforms other recent autoregressive motion prediction methods: MotionLM and StateTransformer, which demonstrates the effectiveness of the proposed designs.
☆ Robotics meets Fluid Dynamics: A Characterization of the Induced Airflow around a Quadrotor
The widespread adoption of quadrotors for diverse applications, from agriculture to public safety, necessitates an understanding of the aerodynamic disturbances they create. This paper introduces a computationally lightweight model for estimating the time-averaged magnitude of the induced flow below quadrotors in hover. Unlike related approaches that rely on expensive computational fluid dynamics (CFD) simulations or time-consuming empirical measurements, our method leverages classical theory from turbulent flows. By analyzing over 9 hours of flight data from drones of varying sizes within a large motion capture system, we show that the combined flow from all propellers of the drone is well-approximated by a turbulent jet. Through the use of a novel normalization and scaling, we have developed and experimentally validated a unified model that describes the mean velocity field of the induced flow for different drone sizes. The model accurately describes the far-field airflow in a very large volume below the drone which is difficult to simulate in CFD. Our model, which requires only the drone's mass, propeller size, and drone size for calculations, offers a practical tool for dynamic planning in multi-agent scenarios, ensuring safer operations near humans and optimizing sensor placements.
comment: 7+1 pages
☆ Workload Estimation for Unknown Tasks: A Survey of Machine Learning Under Distribution Shift
Human-robot teams involve humans and robots collaborating to achieve tasks under various environmental conditions. Successful teaming will require robots to adapt autonomously to a human teammate's internal state. An important element of such adaptation is the ability to estimate the human teammates' workload in unknown situations. Existing workload models use machine learning to model the relationships between physiological metrics and workload; however, these methods are susceptible to individual differences and are heavily influenced by other factors. These methods cannot generalize to unknown tasks, as they rely on standard machine learning approaches that assume data consists of independent and identically distributed (IID) samples. This assumption does not necessarily hold for estimating workload for new tasks. A survey of non-IID machine learning techniques is presented, where commonly used techniques are evaluated using three criteria: portability, model complexity, and adaptability. These criteria are used to argue which techniques are most applicable for estimating workload for unknown tasks in dynamic, real-time environments.
☆ Multi-Robot Connected Fermat Spiral Coverage ICAPS24
We introduce the Multi-Robot Connected Fermat Spiral (MCFS), a novel algorithmic framework for Multi-Robot Coverage Path Planning (MCPP) that adapts Connected Fermat Spiral (CFS) from the computer graphics community to multi-robot coordination for the first time. MCFS uniquely enables the orchestration of multiple robots to generate coverage paths that contour around arbitrarily shaped obstacles, a feature that is notably lacking in traditional methods. Our framework not only enhances area coverage and optimizes task performance, particularly in terms of makespan, for workspaces rich in irregular obstacles but also addresses the challenges of path continuity and curvature critical for non-holonomic robots by generating smooth paths without decomposing the workspace. MCFS solves MCPP by constructing a graph of isolines and transforming MCPP into a combinatorial optimization problem, aiming to minimize the makespan while covering all vertices. Our contributions include developing a unified CFS version for scalable and adaptable MCPP, extending it to MCPP with novel optimization techniques for cost reduction and path continuity and smoothness, and demonstrating through extensive experiments that MCFS outperforms existing MCPP methods in makespan, path curvature, coverage ratio, and overlapping ratio. Our research marks a significant step in MCPP, showcasing the fusion of computer graphics and automated planning principles to advance the capabilities of multi-robot systems in complex environments. Our code is available at https://github.com/reso1/MCFS.
comment: accepted to ICAPS24
☆ POLICEd RL: Learning Closed-Loop Robot Control Policies with Provable Satisfaction of Hard Constraints
In this paper, we seek to learn a robot policy guaranteed to satisfy state constraints. To encourage constraint satisfaction, existing RL algorithms typically rely on Constrained Markov Decision Processes and discourage constraint violations through reward shaping. However, such soft constraints cannot offer verifiable safety guarantees. To address this gap, we propose POLICEd RL, a novel RL algorithm explicitly designed to enforce affine hard constraints in closed-loop with a black-box environment. Our key insight is to force the learned policy to be affine around the unsafe set and use this affine region as a repulsive buffer to prevent trajectories from violating the constraint. We prove that such policies exist and guarantee constraint satisfaction. Our proposed framework is applicable to both systems with continuous and discrete state and action spaces and is agnostic to the choice of the RL training algorithm. Our results demonstrate the capacity of POLICEd RL to enforce hard constraints in robotic tasks while significantly outperforming existing methods.
comment: 26 pages, 11 figures
☆ Map-Aware Human Pose Prediction for Robot Follow-Ahead
In the robot follow-ahead task, a mobile robot is tasked to maintain its relative position in front of a moving human actor while keeping the actor in sight. To accomplish this task, it is important that the robot understand the full 3D pose of the human (since the head orientation can be different than the torso) and predict future human poses so as to plan accordingly. This prediction task is especially tricky in a complex environment with junctions and multiple corridors. In this work, we address the problem of forecasting the full 3D trajectory of a human in such environments. Our main insight is to show that one can first predict the 2D trajectory and then estimate the full 3D trajectory by conditioning the estimator on the predicted 2D trajectory. With this approach, we achieve results comparable or better than the state-of-the-art methods three times faster. As part of our contribution, we present a new dataset where, in contrast to existing datasets, the human motion is in a much larger area than a single room. We also present a complete robot system that integrates our human pose forecasting network on the mobile robot to enable real-time robot follow-ahead and present results from real-world experiments in multiple buildings on campus. Our project page, including supplementary material and videos, can be found at: https://qingyuan-jiang.github.io/iros2024_poseForecasting/
☆ Look Before You Leap: Socially Acceptable High-Speed Ground Robot Navigation in Crowded Hallways IROS 2024
To operate safely and efficiently, autonomous warehouse/delivery robots must be able to accomplish tasks while navigating in dynamic environments and handling the large uncertainties associated with the motions/behaviors of other robots and/or humans. A key scenario in such environments is the hallway problem, where robots must operate in the same narrow corridor as human traffic going in one or both directions. Traditionally, robot planners have tended to focus on socially acceptable behavior in the hallway scenario at the expense of performance. This paper proposes a planner that aims to address the consequent "robot freezing problem" in hallways by allowing for "peek-and-pass" maneuvers. We then go on to demonstrate in simulation how this planner improves robot time to goal without violating social norms. Finally, we show initial hardware demonstrations of this planner in the real world.
comment: Submitted to IROS 2024
☆ Waypoint-Based Reinforcement Learning for Robot Manipulation Tasks
Robot arms should be able to learn new tasks. One framework here is reinforcement learning, where the robot is given a reward function that encodes the task, and the robot autonomously learns actions to maximize its reward. Existing approaches to reinforcement learning often frame this problem as a Markov decision process, and learn a policy (or a hierarchy of policies) to complete the task. These policies reason over hundreds of fine-grained actions that the robot arm needs to take: e.g., moving slightly to the right or rotating the end-effector a few degrees. But the manipulation tasks that we want robots to perform can often be broken down into a small number of high-level motions: e.g., reaching an object or turning a handle. In this paper we therefore propose a waypoint-based approach for model-free reinforcement learning. Instead of learning a low-level policy, the robot now learns a trajectory of waypoints, and then interpolates between those waypoints using existing controllers. Our key novelty is framing this waypoint-based setting as a sequence of multi-armed bandits: each bandit problem corresponds to one waypoint along the robot's motion. We theoretically show that an ideal solution to this reformulation has lower regret bounds than standard frameworks. We also introduce an approximate posterior sampling solution that builds the robot's motion one waypoint at a time. Results across benchmark simulations and two real-world experiments suggest that this proposed approach learns new tasks more quickly than state-of-the-art baselines. See videos here: https://youtu.be/MMEd-lYfq4Y
☆ UNO Push: Unified Nonprehensile Object Pushing via Non-Parametric Estimation and Model Predictive Control
Nonprehensile manipulation through precise pushing is an essential skill that has been commonly challenged by perception and physical uncertainties, such as those associated with contacts, object geometries, and physical properties. For this, we propose a unified framework that jointly addresses system modeling, action generation, and control. While most existing approaches either heavily rely on a priori system information for analytic modeling, or leverage a large dataset to learn dynamic models, our framework approximates a system transition function via non-parametric learning only using a small number of exploratory actions (ca. 10). The approximated function is then integrated with model predictive control to provide precise pushing manipulation. Furthermore, we show that the approximated system transition functions can be robustly transferred across novel objects while being online updated to continuously improve the manipulation accuracy. Through extensive experiments on a real robot platform with a set of novel objects and comparing against a state-of-the-art baseline, we show that the proposed unified framework is a light-weight and highly effective approach to enable precise pushing manipulation all by itself. Our evaluation results illustrate that the system can robustly ensure millimeter-level precision and can straightforwardly work on any novel object.
☆ Enhancing Security in Multi-Robot Systems through Co-Observation Planning, Reachability Analysis, and Network Flow
This paper addresses security challenges in multi-robot systems (MRS) where adversaries may compromise robot control, risking unauthorized access to forbidden areas. We propose a novel multi-robot optimal planning algorithm that integrates mutual observations and introduces reachability constraints for enhanced security. This ensures that, even with adversarial movements, compromised robots cannot breach forbidden regions without missing scheduled co-observations. The reachability constraint uses ellipsoidal over-approximation for efficient intersection checking and gradient computation. To enhance system resilience and tackle feasibility challenges, we also introduce sub-teams. These cohesive units replace individual robot assignments along each route, enabling redundant robots to deviate for co-observations across different trajectories, securing multiple sub-teams without requiring modifications. We formulate the cross-trajectory co-observation plan by solving a network flow coverage problem on the checkpoint graph generated from the original unsecured MRS trajectories, providing the same security guarantees against plan-deviation attacks. We demonstrate the effectiveness and robustness of our proposed algorithm, which significantly strengthens the security of multi-robot systems in the face of adversarial threats.
comment: 12 pages, 6 figures, submitted to IEEE Transactions on Control of Network Systems
☆ A Rule-Compliance Path Planner for Lane-Merge Scenarios Based on Responsibility-Sensitive Safety IROS 2024
Lane merging is one of the critical tasks for self-driving cars, and how to perform lane-merge maneuvers effectively and safely has become one of the important standards in measuring the capability of autonomous driving systems. However, due to the ambiguity in driving intentions and right-of-way issues, the lane merging process in autonomous driving remains deficient in terms of maintaining or ceding the right-of-way and attributing liability, which could result in protracted durations for merging and problems such as trajectory oscillation. Hence, we present a rule-compliance path planner (RCPP) for lane-merge scenarios, which initially employs the extended responsibility-sensitive safety (RSS) to elucidate the right-of-way, followed by the potential field-based sigmoid planner for path generation. In the simulation, we have validated the efficacy of the proposed algorithm. The algorithm demonstrated superior performance over previous approaches in aspects such as merging time (Saved 72.3%), path length (reduced 53.4%), and eliminating the trajectory oscillation.
comment: Submitted to IEEE IROS 2024
☆ Federated reinforcement learning for robot motion planning with zero-shot generalization
This paper considers the problem of learning a control policy for robot motion planning with zero-shot generalization, i.e., no data collection and policy adaptation is needed when the learned policy is deployed in new environments. We develop a federated reinforcement learning framework that enables collaborative learning of multiple learners and a central server, i.e., the Cloud, without sharing their raw data. In each iteration, each learner uploads its local control policy and the corresponding estimated normalized arrival time to the Cloud, which then computes the global optimum among the learners and broadcasts the optimal policy to the learners. Each learner then selects between its local control policy and that from the Cloud for next iteration. The proposed framework leverages on the derived zero-shot generalization guarantees on arrival time and safety. Theoretical guarantees on almost-sure convergence, almost consensus, Pareto improvement and optimality gap are also provided. Monte Carlo simulation is conducted to evaluate the proposed framework.
☆ AMCO: Adaptive Multimodal Coupling of Vision and Proprioception for Quadruped Robot Navigation in Outdoor Environments
We present AMCO, a novel navigation method for quadruped robots that adaptively combines vision-based and proprioception-based perception capabilities. Our approach uses three cost maps: general knowledge map; traversability history map; and current proprioception map; which are derived from a robot's vision and proprioception data, and couples them to obtain a coupled traversability cost map for navigation. The general knowledge map encodes terrains semantically segmented from visual sensing, and represents a terrain's typically expected traversability. The traversability history map encodes the robot's recent proprioceptive measurements on a terrain and its semantic segmentation as a cost map. Further, the robot's present proprioceptive measurement is encoded as a cost map in the current proprioception map. As the general knowledge map and traversability history map rely on semantic segmentation, we evaluate the reliability of the visual sensory data by estimating the brightness and motion blur of input RGB images and accordingly combine the three cost maps to obtain the coupled traversability cost map used for navigation. Leveraging this adaptive coupling, the robot can depend on the most reliable input modality available. Finally, we present a novel planner that selects appropriate gaits and velocities for traversing challenging outdoor environments using the coupled traversability cost map. We demonstrate AMCO's navigation performance in different real-world outdoor environments and observe 10.8%-34.9% reduction w.r.t. two stability metrics, and up to 50% improvement in terms of success rate compared to current navigation methods.
comment: 8 pages
☆ A Contact Model based on Denoising Diffusion to Learn Variable Impedance Control for Contact-rich Manipulation
In this paper, a novel approach is proposed for learning robot control in contact-rich tasks such as wiping, by developing Diffusion Contact Model (DCM). Previous methods of learning such tasks relied on impedance control with time-varying stiffness tuning by performing Bayesian optimization by trial-and-error with robots. The proposed approach aims to reduce the cost of robot operation by predicting the robot contact trajectories from the variable stiffness inputs and using neural models. However, contact dynamics are inherently highly nonlinear, and their simulation requires iterative computations such as convex optimization. Moreover, approximating such computations by using finite-layer neural models is difficult. To overcome these limitations, the proposed DCM used the denoising diffusion models that could simulate the complex dynamics via iterative computations of multi-step denoising, thus improving the prediction accuracy. Stiffness tuning experiments conducted in simulated and real environments showed that the DCM achieved comparable performance to a conventional robot-based optimization method while reducing the number of robot trials.
☆ "It's Not a Replacement:" Enabling Parent-Robot Collaboration to Support In-Home Learning Experiences of Young Children
Learning companion robots for young children are increasingly adopted in informal learning environments. Although parents play a pivotal role in their children's learning, very little is known about how parents prefer to incorporate robots into their children's learning activities. We developed prototype capabilities for a learning companion robot to deliver educational prompts and responses to parent-child pairs during reading sessions and conducted in-home user studies involving 10 families with children aged 3-5. Our data indicates that parents want to work with robots as collaborators to augment parental activities to foster children's learning, introducing the notion of parent-robot collaboration. Our findings offer an empirical understanding of the needs and challenges of parent-child interaction in informal learning scenarios and design opportunities for integrating a companion robot into these interactions. We offer insights into how robots might be designed to facilitate parent-robot collaboration, including parenting policies, collaboration patterns, and interaction paradigms.
☆ Quadcopter Team Configurable Motion Guided by a Quadruped
The paper focuses on modeling and experimental evaluation of a quadcopter team configurable coordination guided by a single quadruped robot. We consider the quadcopter team as particles of a two-dimensional deformable body and propose a two-dimensional affine transformation model for safe and collision-free configurable coordination of this heterogeneous robotic system. The proposed affine transformation is decomposed into translation, that is specified by the quadruped global position, and configurable motion of the quadcopters, which is determined by a nonsingular Jacobian matrix so that the quadcopter team can safely navigate a constrained environment while avoiding collision. We propose two methods to experimentally evaluate the proposed heterogeneous robot coordination model. The first method measures real positions of quadcopters, quadruped, and environmental objects all with respect to the global coordinate system. On the other hand, the second method measures position with respect to the local coordinate system fixed on the dog robot which in turn enables safe planning the Jacobian matrix of the quadcopter team while the world is virtually approached the robotic system.
☆ HRI Curriculum for a Liberal Arts Education
In this paper, we discuss the opportunities and challenges of teaching a human-robot interaction course at an undergraduate liberal arts college. We provide a sample syllabus adapted from a previous version of a course.
comment: Presented at the Designing an Intro to HRI Course Workshop at HRI 2024 (arXiv:2403.05588)
☆ Crowdsourcing Task Traces for Service Robotics
Demonstration is an effective end-user development paradigm for teaching robots how to perform new tasks. In this paper, we posit that demonstration is useful not only as a teaching tool, but also as a way to understand and assist end-user developers in thinking about a task at hand. As a first step toward gaining this understanding, we constructed a lightweight web interface to crowdsource step-by-step instructions of common household tasks, leveraging the imaginations and past experiences of potential end-user developers. As evidence of the utility of our interface, we deployed the interface on Amazon Mechanical Turk and collected 207 task traces that span 18 different task categories. We describe our vision for how these task traces can be operationalized as task models within end-user development tools and provide a roadmap for future work.
comment: Published in the companion proceedings of the 2023 ACM/IEEE International Conference on Human-Robot Interaction
☆ Visual Imitation Learning of Task-Oriented Object Grasping and Rearrangement
Task-oriented object grasping and rearrangement are critical skills for robots to accomplish different real-world manipulation tasks. However, they remain challenging due to partial observations of the objects and shape variations in categorical objects. In this paper, we propose the Multi-feature Implicit Model (MIMO), a novel object representation that encodes multiple spatial features between a point and an object in an implicit neural field. Training such a model on multiple features ensures that it embeds the object shapes consistently in different aspects, thus improving its performance in object shape reconstruction from partial observation, shape similarity measure, and modeling spatial relations between objects. Based on MIMO, we propose a framework to learn task-oriented object grasping and rearrangement from single or multiple human demonstration videos. The evaluations in simulation show that our approach outperforms the state-of-the-art methods for multi- and single-view observations. Real-world experiments demonstrate the efficacy of our approach in one- and few-shot imitation learning of manipulation tasks.
☆ Goal-Oriented End-User Programming of Robots
End-user programming (EUP) tools must balance user control with the robot's ability to plan and act autonomously. Many existing task-oriented EUP tools enforce a specific level of control, e.g., by requiring that users hand-craft detailed sequences of actions, rather than offering users the flexibility to choose the level of task detail they wish to express. We thereby created a novel EUP system, Polaris, that in contrast to most existing EUP tools, uses goal predicates as the fundamental building block of programs. Users can thereby express high-level robot objectives or lower-level checkpoints at their choosing, while an off-the-shelf task planner fills in any remaining program detail. To ensure that goal-specified programs adhere to user expectations of robot behavior, Polaris is equipped with a Plan Visualizer that exposes the planner's output to the user before runtime. In what follows, we describe our design of Polaris and its evaluation with 32 human participants. Our results support the Plan Visualizer's ability to help users craft higher-quality programs. Furthermore, there are strong associations between user perception of the robot and Plan Visualizer usage, and evidence that robot familiarity has a key role in shaping user experience.
comment: Published in the proceedings of the 2024 ACM/IEEE International Conference on Human-Robot Interaction
☆ Open Access NAO (OAN): a ROS2-based software framework for HRI applications with the NAO robot
This paper presents a new software framework for HRI experimentation with the sixth version of the common NAO robot produced by the United Robotics Group. Embracing the common demand of researchers for better performance and new features for NAO, the authors took advantage of the ability to run ROS2 onboard on the NAO to develop a framework independent of the APIs provided by the manufacturer. Such a system provides NAO with not only the basic skills of a humanoid robot such as walking and reproducing movements of interest but also features often used in HRI such as: speech recognition/synthesis, face and object detention, and the use of Generative Pre-trained Transformer (GPT) models for conversation. The developed code is therefore configured as a ready-to-use but also highly expandable and improvable tool thanks to the possibilities provided by the ROS community.
comment: 7 pages, 3 figures
☆ Sensory Glove-Based Surgical Robot User Interface IROS
Robotic surgery has reached a high level of maturity and has become an integral part of standard surgical care. However, existing surgeon consoles are bulky and take up valuable space in the operating room, present challenges for surgical team coordination, and their proprietary nature makes it difficult to take advantage of recent technological advances, especially in virtual and augmented reality. One potential area for further improvement is the integration of modern sensory gloves into robotic platforms, allowing surgeons to control robotic arms directly with their hand movements intuitively. We propose one such system that combines an HTC Vive tracker, a Manus Meta Prime 3 XR sensory glove, and God Vision wireless smart glasses. The system controls one arm of a da Vinci surgical robot. In addition to moving the arm, the surgeon can use fingers to control the end-effector of the surgical instrument. Hand gestures are used to implement clutching and similar functions. In particular, we introduce clutching of the instrument orientation, a functionality not available in the da Vinci system. The vibrotactile elements of the glove are used to provide feedback to the user when gesture commands are invoked. A preliminary evaluation of the system shows that it has excellent tracking accuracy and allows surgeons to efficiently perform common surgical training tasks with minimal practice with the new interface; this suggests that the interface is highly intuitive. The proposed system is inexpensive, allows rapid prototyping, and opens opportunities for further innovations in the design of surgical robot interfaces.
comment: 6 pages, 5 figures, 7 tables, submitted to International Conference on Intelligent Robots and Systems (IROS)2024
☆ Safety-Aware Perception for Autonomous Collision Avoidance in Dynamic Environments
Autonomous collision avoidance requires accurate environmental perception; however, flight systems often possess limited sensing capabilities with field-of-view (FOV) restrictions. To navigate this challenge, we present a safety-aware approach for online determination of the optimal sensor-pointing direction $\psi_\text{d}$ which utilizes control barrier functions (CBFs). First, we generate a spatial density function $\Phi$ which leverages CBF constraints to map the collision risk of all local coordinates. Then, we convolve $\Phi$ with an attitude-dependent sensor FOV quality function to produce the objective function $\Gamma$ which quantifies the total observed risk for a given pointing direction. Finally, by finding the global optimizer for $\Gamma$, we identify the value of $\psi_\text{d}$ which maximizes the perception of risk within the FOV. We incorporate $\psi_\text{d}$ into a safety-critical flight architecture and conduct a numerical analysis using multiple simulated mission profiles. Our algorithm achieves a success rate of $88-96\%$, constituting a $16-29\%$ improvement compared to the best heuristic methods. We demonstrate the functionality of our approach via a flight demonstration using the Crazyflie 2.1 micro-quadrotor. Without a priori obstacle knowledge, the quadrotor follows a dynamic flight path while simultaneously calculating and tracking $\psi_\text{d}$ to perceive and avoid two static obstacles with an average computation time of 371 $\mu$s.
☆ Augmented Reality Demonstrations for Scalable Robot Imitation Learning
Robot Imitation Learning (IL) is a widely used method for training robots to perform manipulation tasks that involve mimicking human demonstrations to acquire skills. However, its practicality has been limited due to its requirement that users be trained in operating real robot arms to provide demonstrations. This paper presents an innovative solution: an Augmented Reality (AR)-assisted framework for demonstration collection, empowering non-roboticist users to produce demonstrations for robot IL using devices like the HoloLens 2. Our framework facilitates scalable and diverse demonstration collection for real-world tasks. We validate our approach with experiments on three classical robotics tasks: reach, push, and pick-and-place. The real robot performs each task successfully while replaying demonstrations collected via AR.
♻ ☆ Ada-NAV: Adaptive Trajectory Length-Based Sample Efficient Policy Learning for Robotic Navigation
Trajectory length stands as a crucial hyperparameter within reinforcement learning (RL) algorithms, significantly contributing to the sample inefficiency in robotics applications. Motivated by the pivotal role trajectory length plays in the training process, we introduce Ada-NAV, a novel adaptive trajectory length scheme designed to enhance the training sample efficiency of RL algorithms in robotic navigation tasks. Unlike traditional approaches that treat trajectory length as a fixed hyperparameter, we propose to dynamically adjust it based on the entropy of the underlying navigation policy. Interestingly, Ada-NAV can be applied to both existing on-policy and off-policy RL methods, which we demonstrate by empirically validating its efficacy on three popular RL methods: REINFORCE, Proximal Policy Optimization (PPO), and Soft Actor-Critic (SAC). We demonstrate through simulated and real-world robotic experiments that Ada-NAV outperforms conventional methods that employ constant or randomly sampled trajectory lengths. Specifically, for a fixed sample budget, Ada-NAV achieves an 18\% increase in navigation success rate, a 20-38\% reduction in navigation path length, and a 9.32\% decrease in elevation costs. Furthermore, we showcase the versatility of Ada-NAV by integrating it with the Clearpath Husky robot, illustrating its applicability in complex outdoor environments.
comment: 11 pages, 9 figures, 2 tables
♻ ☆ uPLAM: Robust Panoptic Localization and Mapping Leveraging Perception Uncertainties
The availability of a robust map-based localization system is essential for the operation of many autonomously navigating vehicles. Since uncertainty is an inevitable part of perception, it is beneficial for the robustness of the robot to consider it in typical downstream tasks of navigation stacks. In particular localization and mapping methods, which in modern systems often employ convolutional neural networks (CNNs) for perception tasks, require proper uncertainty estimates. In this work, we present uncertainty-aware Panoptic Localization and Mapping (uPLAM), which employs pixel-wise uncertainty estimates for panoptic CNNs as a bridge to fuse modern perception with classical probabilistic localization and mapping approaches. Beyond the perception, we introduce an uncertainty-based map aggregation technique to create accurate panoptic maps, containing surface semantics and landmark instances. Moreover, we provide cell-wise map uncertainties, and present a particle filter-based localization method that employs perception uncertainties. Extensive evaluations show that our proposed incorporation of uncertainties leads to more accurate maps with reliable uncertainty estimates and improved localization accuracy. Additionally, we present the Freiburg Panoptic Driving dataset for evaluating panoptic mapping and localization methods. We make our code and dataset available at: \url{http://uplam.cs.uni-freiburg.de}
♻ ☆ Exposing the Unseen: Exposure Time Emulation for Offline Benchmarking of Vision Algorithms IROS 2024
Visual Odometry (VO) is one of the fundamental tasks in computer vision for robotics. However, its performance is deeply affected by High Dynamic Range (HDR) scenes, omnipresent outdoor. While new Automatic-Exposure (AE) approaches to mitigate this have appeared, their comparison in a reproducible manner is problematic. This stems from the fact that the behavior of AE depends on the environment, and it affects the image acquisition process. Consequently, AE has traditionally only been benchmarked in an online manner, making the experiments non-reproducible. To solve this, we propose a new methodology based on an emulator that can generate images at any exposure time. It leverages BorealHDR, a unique multi-exposure stereo dataset collected over 10 km, on 55 trajectories with challenging illumination conditions. Moreover, it includes lidar-inertial-based global maps with pose estimation for each image frame as well as Global Navigation Satellite System (GNSS) data, for comparison. We show that using these images acquired at different exposure times, we can emulate realistic images, keeping a Root-Mean-Square Error (RMSE) below 1.78 % compared to ground truth images. To demonstrate the practicality of our approach for offline benchmarking, we compared three state-of-the-art AE algorithms on key elements of Visual Simultaneous Localization And Mapping (VSLAM) pipeline, against four baselines. Consequently, reproducible evaluation of AE is now possible, speeding up the development of future approaches. Our code and dataset are available online at this link: https://github.com/norlab-ulaval/BorealHDR
comment: 8 pages, 6 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)
♻ ☆ Investigation of Enhanced Inertial Navigation Algorithms by Functional Iteration
The defects of the traditional strapdown inertial navigation algorithms become well acknowledged and the corresponding enhanced algorithms have been quite recently proposed trying to mitigate both theoretical and algorithmic defects. In this paper, the analytical accuracy evaluation of both the traditional algorithms and the enhanced algorithms is investigated, against the true reference for the first time enabled by the functional iteration approach having provable convergence. The analyses by the help of MATLAB Symbolic Toolbox show that the resultant error orders of all algorithms under investigation are consistent with those in the existing literatures, and the enhanced attitude algorithm notably reduces error orders of the traditional counterpart, while the impact of the enhanced velocity algorithm on error order reduction is insignificant. Simulation results agree with analyses that the superiority of the enhanced algorithm over the traditional one in the body-frame attitude computation scenario diminishes significantly in the entire inertial navigation computation scenario, while the functional iteration approach possesses significant accuracy superiority even under sustained lowly dynamic conditions.
comment: 12 pages, 3 figs
♻ ☆ Intention-Aware Planner for Robust and Safe Aerial Tracking IROS
Autonomous target tracking with quadrotors has wide applications in many scenarios, such as cinematographic follow-up shooting or suspect chasing. Target motion prediction is necessary when designing the tracking planner. However, the widely used constant velocity or constant rotation assumption can not fully capture the dynamics of the target. The tracker may fail when the target happens to move aggressively, such as sudden turn or deceleration. In this paper, we propose an intention-aware planner by additionally considering the intention of the target to enhance safety and robustness in aerial tracking applications. Firstly, a designated intention prediction method is proposed, which combines a user-defined potential assessment function and a state observation function. A reachable region is generated to specifically evaluate the turning intentions. Then we design an intention-driven hybrid A* method to predict the future possible positions for the target. Finally, an intention-aware optimization approach is designed to generate a spatial-temporal optimal trajectory, allowing the tracker to perceive unexpected situations from the target. Benchmark comparisons and real-world experiments are conducted to validate the performance of our method.
comment: 8 pages, 10 figures, submitted to 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
♻ ☆ Surfer: Progressive Reasoning with World Models for Robotic Manipulation
Considering how to make the model accurately understand and follow natural language instructions and perform actions consistent with world knowledge is a key challenge in robot manipulation. This mainly includes human fuzzy instruction reasoning and the following of physical knowledge. Therefore, the embodied intelligence agent must have the ability to model world knowledge from training data. However, most existing vision and language robot manipulation methods mainly operate in less realistic simulator and language settings and lack explicit modeling of world knowledge. To bridge this gap, we introduce a novel and simple robot manipulation framework, called Surfer. It is based on the world model, treats robot manipulation as a state transfer of the visual scene, and decouples it into two parts: action and scene. Then, the generalization ability of the model on new instructions and new scenes is enhanced by explicit modeling of the action and scene prediction in multi-modal information. In addition to the framework, we also built a robot manipulation simulator that supports full physics execution based on the MuJoCo physics engine. It can automatically generate demonstration training data and test data, effectively reducing labor costs. To conduct a comprehensive and systematic evaluation of the robot manipulation model in terms of language understanding and physical execution, we also created a robotic manipulation benchmark with progressive reasoning tasks, called SeaWave. It contains 4 levels of progressive reasoning tasks and can provide a standardized testing platform for embedded AI agents in multi-modal environments. On average, Surfer achieved a success rate of 54.74% on the defined four levels of manipulation tasks, exceeding the best baseline performance of 47.64%.
♻ ☆ LLM3:Large Language Model-based Task and Motion Planning with Motion Failure Reasoning IROS 2024
Conventional Task and Motion Planning (TAMP) approaches rely on manually crafted interfaces connecting symbolic task planning with continuous motion generation. These domain-specific and labor-intensive modules are limited in addressing emerging tasks in real-world settings. Here, we present LLM^3, a novel Large Language Model (LLM)-based TAMP framework featuring a domain-independent interface. Specifically, we leverage the powerful reasoning and planning capabilities of pre-trained LLMs to propose symbolic action sequences and select continuous action parameters for motion planning. Crucially, LLM^3 incorporates motion planning feedback through prompting, allowing the LLM to iteratively refine its proposals by reasoning about motion failure. Consequently, LLM^3 interfaces between task planning and motion planning, alleviating the intricate design process of handling domain-specific messages between them. Through a series of simulations in a box-packing domain, we quantitatively demonstrate the effectiveness of LLM^3 in solving TAMP problems and the efficiency in selecting action parameters. Ablation studies underscore the significant contribution of motion failure reasoning to the success of LLM^3. Furthermore, we conduct qualitative experiments on a physical manipulator, demonstrating the practical applicability of our approach in real-world settings.
comment: Submitted to IROS 2024. Codes available: https://github.com/AssassinWS/LLM-TAMP
♻ ☆ Pseudo-rigid body networks: learning interpretable deformable object dynamics from partial observations
Accurate prediction of deformable linear object (DLO) dynamics is challenging if the task at hand requires a human-interpretable yet computationally fast model. In this work, we draw inspiration from the pseudo-rigid body method (PRB) and model a DLO as a serial chain of rigid bodies whose internal state is unrolled through time by a dynamics network. This dynamics network is trained jointly with a physics-informed encoder which maps observed motion variables to the DLO's hidden state. To encourage that the state acquires a physically meaningful representation, we leverage the forward kinematics of the PRB model as decoder. We demonstrate in robot experiments that the proposed DLO dynamics model provides physically interpretable predictions from partial observations while being on par with black-box models regarding prediction accuracy. The project code is available at: http://tinyurl.com/prb-networks
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
♻ ☆ Camera Height Doesn't Change: Unsupervised Training for Metric Monocular Road-Scene Depth Estimation
In this paper, we introduce a novel training method for making any monocular depth network learn absolute scale and estimate metric road-scene depth just from regular training data, i.e., driving videos. We refer to this training framework as StableCamH. The key idea is to leverage cars found on the road as sources of scale supervision but to incorporate them in the training robustly. StableCamH detects and estimates the sizes of cars in the frame and aggregates scale information extracted from them into a camera height estimate whose consistency across the entire video sequence is enforced as scale supervision. This realizes robust unsupervised training of any, otherwise scale-oblivious, monocular depth network to become not only scale-aware but also metric-accurate without the need for auxiliary sensors and extra supervision. Extensive experiments on the KITTI and Cityscapes datasets show the effectiveness of StableCamH and its state-of-the-art accuracy compared with related methods. We also show that StableCamH enables training on mixed datasets of different camera heights, which leads to larger-scale training and thus higher generalization. Metric depth reconstruction is essential in any road-scene visual modeling, and StableCamH democratizes its deployment by establishing the means to train any model as a metric depth estimator.
♻ ☆ NDT-Map-Code: A 3D global descriptor for real-time loop closure detection in lidar SLAM
Loop-closure detection, also known as place recognition, aiming to identify previously visited locations, is an essential component of a SLAM system. Existing research on lidar-based loop closure heavily relies on dense point cloud and 360 FOV lidars. This paper proposes an out-of-the-box NDT (Normal Distribution Transform) based global descriptor, NDT-Map-Code, designed for both on-road driving and underground valet parking scenarios. NDT-Map-Code can be directly extracted from the NDT map without the need for a dense point cloud, resulting in excellent scalability and low maintenance cost. The NDT representation is leveraged to identify representative patterns, which are further encoded according to their spatial location (bearing, range, and height). Experimental results on the NIO underground parking lot dataset and the KITTI dataset demonstrate that our method achieves significantly better performance compared to the state-of-the-art.
comment: 8 pages, 6 figures, 4 tables
♻ ☆ Vision-State Fusion: Improving Deep Neural Networks for Autonomous Robotics
Vision-based deep learning perception fulfills a paramount role in robotics, facilitating solutions to many challenging scenarios, such as acrobatic maneuvers of autonomous unmanned aerial vehicles (UAVs) and robot-assisted high-precision surgery. Control-oriented end-to-end perception approaches, which directly output control variables for the robot, commonly take advantage of the robot's state estimation as an auxiliary input. When intermediate outputs are estimated and fed to a lower-level controller, i.e. mediated approaches, the robot's state is commonly used as an input only for egocentric tasks, which estimate physical properties of the robot itself. In this work, we propose to apply a similar approach for the first time -- to the best of our knowledge -- to non-egocentric mediated tasks, where the estimated outputs refer to an external subject. We prove how our general methodology improves the regression performance of deep convolutional neural networks (CNNs) on a broad class of non-egocentric 3D pose estimation problems, with minimal computational cost. By analyzing three highly-different use cases, spanning from grasping with a robotic arm to following a human subject with a pocket-sized UAV, our results consistently improve the R\textsuperscript{2} regression metric, up to +0.51, compared to their stateless baselines. Finally, we validate the in-field performance of a closed-loop autonomous cm-scale UAV on the human pose estimation task. Our results show a significant reduction, i.e., 24\% on average, on the mean absolute error of our stateful CNN, compared to a State-of-the-Art stateless counterpart.
comment: This paper has been accepted for publication in the Journal of Intelligent & Robotic Systems. \copyright 2024 Springer
♻ ☆ Distributed Pose-graph Optimization with Multi-level Partitioning for Collaborative SLAM
The back-end module of Distributed Collaborative Simultaneous Localization and Mapping (DCSLAM) requires solving a nonlinear Pose Graph Optimization (PGO) under a distributed setting, also known as SE(d)-synchronization. Most existing distributed graph optimization algorithms employ a simple sequential partitioning scheme, which may result in unbalanced subgraph dimensions due to the different geographic locations of each robot, and hence imposes extra communication load. Moreover, the performance of current Riemannian optimization algorithms can be further accelerated. In this letter, we propose a novel distributed pose graph optimization algorithm combining multi-level partitioning with an accelerated Riemannian optimization method. Firstly, we employ the multi-level graph partitioning algorithm to preprocess the naive pose graph to formulate a balanced optimization problem. In addition, inspired by the accelerated coordinate descent method, we devise an Improved Riemannian Block Coordinate Descent (IRBCD) algorithm and the critical point obtained is globally optimal. Finally, we evaluate the effects of four common graph partitioning approaches on the correlation of the inter-subgraphs, and discover that the Highest scheme has the best partitioning performance. Also, we implement simulations to quantitatively demonstrate that our proposed algorithm outperforms the state-of-the-art distributed pose graph optimization protocols.
♻ ☆ Results and Lessons Learned from Autonomous Driving Transportation Services in Airfield, Crowded Indoor, and Urban Environments
Autonomous vehicles have been actively investigated over the past few decades. Several recent works show the potential of autonomous vehicles in urban environments with impressive experimental results. However, these works note that autonomous vehicles are still occasionally inferior to expert drivers in complex scenarios. Furthermore, they do not focus on the possibilities of autonomous driving transportation services in other areas beyond urban environments. This paper presents the research results and lessons learned from autonomous driving transportation services in airfield, crowded indoor, and urban environments. We discuss how we address several unique challenges in these diverse environments. We also offer an overview of remaining challenges that have not received much attention but must be addressed. This paper aims to share our unique experience to support researchers who are interested in exploring autonomous driving transportation services in various real-world environments.
comment: 8 pages, 7 figures, 4 tables
♻ ☆ Improving Out-of-Distribution Generalization of Learned Dynamics by Learning Pseudometrics and Constraint Manifolds ICRA 2024
We propose a method for improving the prediction accuracy of learned robot dynamics models on out-of-distribution (OOD) states. We achieve this by leveraging two key sources of structure often present in robot dynamics: 1) sparsity, i.e., some components of the state may not affect the dynamics, and 2) physical limits on the set of possible motions, in the form of nonholonomic constraints. Crucially, we do not assume this structure is known a priori, and instead learn it from data. We use contrastive learning to obtain a distance pseudometric that uncovers the sparsity pattern in the dynamics, and use it to reduce the input space when learning the dynamics. We then learn the unknown constraint manifold by approximating the normal space of possible motions from the data, which we use to train a Gaussian process (GP) representation of the constraint manifold. We evaluate our approach on a physical differential-drive robot and a simulated quadrotor, showing improved prediction accuracy on OOD data relative to baselines.
comment: Accept to ICRA 2024, 6 pages + references
♻ ☆ LingoQA: Video Question Answering for Autonomous Driving
Autonomous driving has long faced a challenge with public acceptance due to the lack of explainability in the decision-making process. Video question-answering (QA) in natural language provides the opportunity for bridging this gap. Nonetheless, evaluating the performance of Video QA models has proved particularly tough due to the absence of comprehensive benchmarks. To fill this gap, we introduce LingoQA, a benchmark specifically for autonomous driving Video QA. The LingoQA trainable metric demonstrates a 0.95 Spearman correlation coefficient with human evaluations. We introduce a Video QA dataset of central London consisting of 419k samples that we release with the paper. We establish a baseline vision-language model and run extensive ablation studies to understand its performance.
comment: Benchmark and dataset are available at https://github.com/wayveai/LingoQA/
♻ ☆ Designing Library of Skill-Agents for Hardware-Level Reusability
To use new robot hardware in a new environment, it is necessary to develop a control program tailored to that specific robot in that environment. Considering the reusability of software among robots is crucial to minimize the effort involved in this process and maximize software reuse across different robots in different environments. This paper proposes a method to remedy this process by considering hardware-level reusability, using Learning-from-observation (LfO) paradigm with a pre-designed skill-agent library. The LfO framework represents the required actions in hardware-independent representations, referred to as task models, from observing human demonstrations, capturing the necessary parameters for the interaction between the environment and the robot. When executing the desired actions from the task models, a set of skill agents is employed to convert the representations into robot commands. This paper focuses on the latter part of the LfO framework, utilizing the set to generate robot actions from the task models, and explores a hardware-independent design approach for these skill agents. These skill agents are described in a hardware-independent manner, considering the relative relationship between the robot's hand position and the environment. As a result, it is possible to execute these actions on robots with different hardware configurations by simply swapping the inverse kinematics solver. This paper, first, defines a necessary and sufficient skill-agent set corresponding to cover all possible actions, and considers the design principles for these skill agents in the library. We provide concrete examples of such skill agents and demonstrate the practicality of using these skill agents by showing that the same representations can be executed on two different robots, Nextage and Fetch, using the proposed skill-agents set.
♻ ☆ Working Backwards: Learning to Place by Picking IROS'24
We present placing via picking (PvP), a method to autonomously collect real-world demonstrations for a family of placing tasks in which objects must be manipulated to specific contact-constrained locations. With PvP, we approach the collection of robotic object placement demonstrations by reversing the grasping process and exploiting the inherent symmetry of the pick and place problems. Specifically, we obtain placing demonstrations from a set of grasp sequences of objects initially located at their target placement locations. Our system can collect hundreds of demonstrations in contact-constrained environments without human intervention by combining two modules: tactile regrasping and compliant control for grasps. We train a policy directly from visual observations through behavioral cloning, using the autonomously-collected demonstrations. By doing so, the policy can generalize to object placement scenarios outside of the training environment without privileged information (e.g., placing a plate picked up from a table). We validate our approach in home robotic scenarios that include dishwasher loading and table setting. Our approach yields robotic placing policies that outperform policies trained with kinesthetic teaching, both in terms of performance and data efficiency, while requiring no human supervision.
comment: Submitted to the IEEE/RSJ International Conference on Intelligent Robotics and Systems (IROS'24), Abu Dhabi, UAE, Oct 14-18, 2024
♻ ☆ PhotoBot: Reference-Guided Interactive Photography via Natural Language IROS'24
We introduce PhotoBot, a framework for fully automated photo acquisition based on an interplay between high-level human language guidance and a robot photographer. We propose to communicate photography suggestions to the user via reference images that are selected from a curated gallery. We leverage a visual language model (VLM) and an object detector to characterize the reference images via textual descriptions and then use a large language model (LLM) to retrieve relevant reference images based on a user's language query through text-based reasoning. To correspond the reference image and the observed scene, we exploit pre-trained features from a vision transformer capable of capturing semantic similarity across marked appearance variations. Using these features, we compute pose adjustments for an RGB-D camera by solving a perspective-n-point (PnP) problem. We demonstrate our approach using a manipulator equipped with a wrist camera. Our user studies show that photos taken by PhotoBot are often more aesthetically pleasing than those taken by users themselves, as measured by human feedback. We also show that PhotoBot can generalize to other reference sources such as paintings.
comment: Submitted to the IEEE/RSJ International Conference on Intelligent Robotics and Systems (IROS'24), Abu Dhabi, UAE, Oct 14-18, 2024
♻ ☆ Social Robots for Sleep Health: A Scoping Review
Poor sleep health is an increasingly concerning public healthcare crisis, especially when coupled with a dwindling number of health professionals qualified to combat it. However, there is a growing body of scientific literature on the use of digital technologies in supporting and sustaining individuals' healthy sleep habits. Social robots are a relatively recent technology that has been used to facilitate health care interventions and may have potential in improving sleep health outcomes, as well. Social robots' unique characteristics -- such as anthropomorphic physical embodiment or effective communication methods -- help to engage users and motivate them to comply with specific interventions, thus improving the interventions' outcomes. This scoping review aims to evaluate current scientific evidence for employing social robots in sleep health interventions, identify critical research gaps, and suggest future directions for developing and using social robots to improve people's sleep health. Our analysis of the reviewed studies found them limited due to a singular focus on the older adult population, use of small sample sizes, limited intervention durations, and other compounding factors. Nevertheless, the reviewed studies reported several positive outcomes, highlighting the potential social robots hold in this field. Although our review found limited clinical evidence for the efficacy of social robots as purveyors of sleep health interventions, it did elucidate the potential for a successful future in this domain if current limitations are addressed and more research is conducted.
♻ ☆ LGMCTS: Language-Guided Monte-Carlo Tree Search for Executable Semantic Object Rearrangement
We introduce a novel approach to the executable semantic object rearrangement problem. In this challenge, a robot seeks to create an actionable plan that rearranges objects within a scene according to a pattern dictated by a natural language description. Unlike existing methods such as StructFormer and StructDiffusion, which tackle the issue in two steps by first generating poses and then leveraging a task planner for action plan formulation, our method concurrently addresses pose generation and action planning. We achieve this integration using a Language-Guided Monte-Carlo Tree Search (LGMCTS). Quantitative evaluations are provided on two simulation datasets, and complemented by qualitative tests with a real robot.
comment: Our code and supplementary materials are accessible at https://github.com/changhaonan/LG-MCTS
Artificial Intelligence 133
☆ On Pretraining Data Diversity for Self-Supervised Learning
We explore the impact of training with more diverse datasets, characterized by the number of unique samples, on the performance of self-supervised learning (SSL) under a fixed computational budget. Our findings consistently demonstrate that increasing pretraining data diversity enhances SSL performance, albeit only when the distribution distance to the downstream data is minimal. Notably, even with an exceptionally large pretraining data diversity achieved through methods like web crawling or diffusion-generated data, among other ways, the distribution shift remains a challenge. Our experiments are comprehensive with seven SSL methods using large-scale datasets such as ImageNet and YFCC100M amounting to over 200 GPU days. Code and trained models will be available at https://github.com/hammoudhasan/DiversitySSL .
comment: Under review
☆ RAR: Retrieving And Ranking Augmented MLLMs for Visual Recognition
CLIP (Contrastive Language-Image Pre-training) uses contrastive learning from noise image-text pairs to excel at recognizing a wide array of candidates, yet its focus on broad associations hinders the precision in distinguishing subtle differences among fine-grained items. Conversely, Multimodal Large Language Models (MLLMs) excel at classifying fine-grained categories, thanks to their substantial knowledge from pre-training on web-level corpora. However, the performance of MLLMs declines with an increase in category numbers, primarily due to growing complexity and constraints of limited context window size. To synergize the strengths of both approaches and enhance the few-shot/zero-shot recognition abilities for datasets characterized by extensive and fine-grained vocabularies, this paper introduces RAR, a Retrieving And Ranking augmented method for MLLMs. We initially establish a multi-modal retriever based on CLIP to create and store explicit memory for different categories beyond the immediate context window. During inference, RAR retrieves the top-k similar results from the memory and uses MLLMs to rank and make the final predictions. Our proposed approach not only addresses the inherent limitations in fine-grained recognition but also preserves the model's comprehensive knowledge base, significantly boosting accuracy across a range of vision-language recognition tasks. Notably, our approach demonstrates a significant improvement in performance on 5 fine-grained visual recognition benchmarks, 11 few-shot image recognition datasets, and the 2 object detection datasets under the zero-shot recognition setting.
comment: Project: https://github.com/Liuziyu77/RAR
☆ ZigMa: Zigzag Mamba Diffusion Model
The diffusion model has long been plagued by scalability and quadratic complexity issues, especially within transformer-based structures. In this study, we aim to leverage the long sequence modeling capability of a State-Space Model called Mamba to extend its applicability to visual data generation. Firstly, we identify a critical oversight in most current Mamba-based vision methods, namely the lack of consideration for spatial continuity in the scan scheme of Mamba. Secondly, building upon this insight, we introduce a simple, plug-and-play, zero-parameter method named Zigzag Mamba, which outperforms Mamba-based baselines and demonstrates improved speed and memory utilization compared to transformer-based baselines. Lastly, we integrate Zigzag Mamba with the Stochastic Interpolant framework to investigate the scalability of the model on large-resolution visual datasets, such as FacesHQ $1024\times 1024$ and UCF101, MultiModal-CelebA-HQ, and MS COCO $256\times 256$. Code will be released at https://taohu.me/zigma/
comment: Project Page: https://taohu.me/zigma/
☆ Natural Language as Polices: Reasoning for Coordinate-Level Embodied Control with LLMs
We demonstrate experimental results with LLMs that address robotics action planning problems. Recently, LLMs have been applied in robotics action planning, particularly using a code generation approach that converts complex high-level instructions into mid-level policy codes. In contrast, our approach acquires text descriptions of the task and scene objects, then formulates action planning through natural language reasoning, and outputs coordinate level control commands, thus reducing the necessity for intermediate representation code as policies. Our approach is evaluated on a multi-modal prompt simulation benchmark, demonstrating that our prompt engineering experiments with natural language reasoning significantly enhance success rates compared to its absence. Furthermore, our approach illustrates the potential for natural language descriptions to transfer robotics skills from known tasks to previously unseen tasks.
comment: 8 pages, 2 figures
☆ Reverse Training to Nurse the Reversal Curse
Large language models (LLMs) have a surprising failure: when trained on "A has a feature B", they do not generalize to "B is a feature of A", which is termed the Reversal Curse. Even when training with trillions of tokens this issue still appears due to Zipf's law - hence even if we train on the entire internet. This work proposes an alternative training scheme, called reverse training, whereby all words are used twice, doubling the amount of available tokens. The LLM is trained in both forward and reverse directions by reversing the training strings while preserving (i.e., not reversing) chosen substrings, such as entities. We show that data-matched reverse-trained models provide superior performance to standard models on standard tasks, and compute-matched reverse-trained models provide far superior performance on reversal tasks, helping resolve the reversal curse issue.
☆ Hierarchical NeuroSymbolic Approach for Action Quality Assessment
Action quality assessment (AQA) applies computer vision to quantitatively assess the performance or execution of a human action. Current AQA approaches are end-to-end neural models, which lack transparency and tend to be biased because they are trained on subjective human judgements as ground-truth. To address these issues, we introduce a neuro-symbolic paradigm for AQA, which uses neural networks to abstract interpretable symbols from video data and makes quality assessments by applying rules to those symbols. We take diving as the case study. We found that domain experts prefer our system and find it more informative than purely neural approaches to AQA in diving. Our system also achieves state-of-the-art action recognition and temporal segmentation, and automatically generates a detailed report that breaks the dive down into its elements and provides objective scoring with visual evidence. As verified by a group of domain experts, this report may be used to assist judges in scoring, help train judges, and provide feedback to divers. We will open-source all of our annotated training data and code for ease of reproducibility.
☆ The Model Openness Framework: Promoting Completeness and Openness for Reproducibility, Transparency and Usability in AI
Generative AI (GAI) offers unprecedented possibilities but its commercialization has raised concerns about transparency, reproducibility, bias, and safety. Many "open-source" GAI models lack the necessary components for full understanding and reproduction, and some use restrictive licenses, a practice known as "openwashing." We propose the Model Openness Framework (MOF), a ranked classification system that rates machine learning models based on their completeness and openness, following principles of open science, open source, open data, and open access. The MOF requires specific components of the model development lifecycle to be included and released under appropriate open licenses. This framework aims to prevent misrepresentation of models claiming to be open, guide researchers and developers in providing all model components under permissive licenses, and help companies, academia, and hobbyists identify models that can be safely adopted without restrictions. Wide adoption of the MOF will foster a more open AI ecosystem, accelerating research, innovation, and adoption.
comment: 45 pages
☆ Information-Theoretic Distillation for Reference-less Summarization
The current winning recipe for automatic summarization is using proprietary large-scale language models (LLMs) such as ChatGPT as is, or imitation learning from them as teacher models. While increasingly ubiquitous dependence on such large-scale language models is convenient, there remains an important question of whether small-scale models could have achieved competitive results, if we were to seek an alternative learning method -- that allows for a more cost-efficient, controllable, yet powerful summarizer. We present InfoSumm, a novel framework to distill a powerful summarizer based on the information-theoretic objective for summarization, without relying on either the LLM's capability or human-written references. To achieve this, we first propose a novel formulation of the desiderata of summarization (saliency, faithfulness and brevity) through the lens of mutual information between the original document and the summary. Based on this formulation, we start off from Pythia-2.8B as the teacher model, which is not yet capable of summarization, then self-train the model to optimize for the information-centric measures of ideal summaries. Distilling from the improved teacher, we arrive at a compact but powerful summarizer with only 568M parameters that performs competitively against ChatGPT, without ever relying on ChatGPT's capabilities. Extensive analysis demonstrates that our approach outperforms in-domain supervised models in human evaluation, let alone state-of-the-art unsupervised methods, and wins over ChatGPT in controllable summarization.
☆ Towards Principled Representation Learning from Videos for Reinforcement Learning ICLR 2024
We study pre-training representations for decision-making using video data, which is abundantly available for tasks such as game agents and software testing. Even though significant empirical advances have been made on this problem, a theoretical understanding remains absent. We initiate the theoretical investigation into principled approaches for representation learning and focus on learning the latent state representations of the underlying MDP using video data. We study two types of settings: one where there is iid noise in the observation, and a more challenging setting where there is also the presence of exogenous noise, which is non-iid noise that is temporally correlated, such as the motion of people or cars in the background. We study three commonly used approaches: autoencoding, temporal contrastive learning, and forward modeling. We prove upper bounds for temporal contrastive learning and forward modeling in the presence of only iid noise. We show that these approaches can learn the latent state and use it to do efficient downstream RL with polynomial sample complexity. When exogenous noise is also present, we establish a lower bound result showing that the sample complexity of learning from video data can be exponentially worse than learning from action-labeled trajectory data. This partially explains why reinforcement learning with video pre-training is hard. We evaluate these representational learning methods in two visual domains, yielding results that are consistent with our theoretical findings.
comment: ICLR 2024 Spotlight Conference Paper
☆ Hyper Strategy Logic AAMAS 2024
Strategy logic (SL) is a powerful temporal logic that enables strategic reasoning in multi-agent systems. SL supports explicit (first-order) quantification over strategies and provides a logical framework to express many important properties such as Nash equilibria, dominant strategies, etc. While in SL the same strategy can be used in multiple strategy profiles, each such profile is evaluated w.r.t. a path-property, i.e., a property that considers the single path resulting from a particular strategic interaction. In this paper, we present Hyper Strategy Logic (HyperSL), a strategy logic where the outcome of multiple strategy profiles can be compared w.r.t. a hyperproperty, i.e., a property that relates multiple paths. We show that HyperSL can capture important properties that cannot be expressed in SL, including non-interference, quantitative Nash equilibria, optimal adversarial planning, and reasoning under imperfect information. On the algorithmic side, we identify an expressive fragment of HyperSL with decidable model checking and present a model-checking algorithm. We contribute a prototype implementation of our algorithm and report on encouraging experimental results.
comment: AAMAS 2024
Reinforcement Learning for Online Testing of Autonomous Driving Systems: a Replication and Extension Study
In a recent study, Reinforcement Learning (RL) used in combination with many-objective search, has been shown to outperform alternative techniques (random search and many-objective search) for online testing of Deep Neural Network-enabled systems. The empirical evaluation of these techniques was conducted on a state-of-the-art Autonomous Driving System (ADS). This work is a replication and extension of that empirical study. Our replication shows that RL does not outperform pure random test generation in a comparison conducted under the same settings of the original study, but with no confounding factor coming from the way collisions are measured. Our extension aims at eliminating some of the possible reasons for the poor performance of RL observed in our replication: (1) the presence of reward components providing contrasting or useless feedback to the RL agent; (2) the usage of an RL algorithm (Q-learning) which requires discretization of an intrinsically continuous state space. Results show that our new RL agent is able to converge to an effective policy that outperforms random testing. Results also highlight other possible improvements, which open to further investigations on how to best leverage RL for online ADS testing.
☆ M-HOF-Opt: Multi-Objective Hierarchical Output Feedback Optimization via Multiplier Induced Loss Landscape Scheduling
When a neural network parameterized loss function consists of many terms, the combinatorial choice of weight multipliers during the optimization process forms a challenging problem. To address this, we proposed a probabilistic graphical model (PGM) for the joint model parameter and multiplier evolution process, with a hypervolume based likelihood that promotes multi-objective descent of each loss term. The corresponding parameter and multiplier estimation as a sequential decision process is then cast into an optimal control problem, where the multi-objective descent goal is dispatched hierarchically into a series of constraint optimization sub-problems. The sub-problem constraint automatically adapts itself according to Pareto dominance and serves as the setpoint for the low level multiplier controller to schedule loss landscapes via output feedback of each loss term. Our method is multiplier-free and operates at the timescale of epochs, thus saves tremendous computational resources compared to full training cycle multiplier tuning. We applied it to domain invariant variational auto-encoding with 6 loss terms on the PACS domain generalization task, and observed robust performance across a range of controller hyperparameters, as well as different multiplier initial conditions, outperforming other multiplier scheduling methods. We offered modular implementation of our method, admitting custom definition of many loss terms for applying our multi-objective hierarchical output feedback training scheme to other deep learning fields.
☆ Large Language Models meet Network Slicing Management and Orchestration
Network slicing, a cornerstone technology for future networks, enables the creation of customized virtual networks on a shared physical infrastructure. This fosters innovation and agility by providing dedicated resources tailored to specific applications. However, current orchestration and management approaches face limitations in handling the complexity of new service demands within multi-administrative domain environments. This paper proposes a future vision for network slicing powered by Large Language Models (LLMs) and multi-agent systems, offering a framework that can be integrated with existing Management and Orchestration (MANO) frameworks. This framework leverages LLMs to translate user intent into technical requirements, map network functions to infrastructure, and manage the entire slice lifecycle, while multi-agent systems facilitate collaboration across different administrative domains. We also discuss the challenges associated with implementing this framework and potential solutions to mitigate them.
☆ Research Re: search & Re-search
Search algorithms are often categorized by their node expansion strategy. One option is the depth-first strategy, a simple backtracking strategy that traverses the search space in the order in which successor nodes are generated. An alternative is the best-first strategy, which was designed to make it possible to use domain-specific heuristic information. By exploring promising parts of the search space first, best-first algorithms are usually more efficient than depth-first algorithms. In programs that play minimax games such as chess and checkers, the efficiency of the search is of crucial importance. Given the success of best-first algorithms in other domains, one would expect them to be used for minimax games too. However, all high-performance game-playing programs are based on a depth-first algorithm. This study takes a closer look at a depth-first algorithm, AB, and a best-first algorithm, SSS. The prevailing opinion on these algorithms is that SSS offers the potential for a more efficient search, but that its complicated formulation and exponential memory requirements render it impractical. The theoretical part of this work shows that there is a surprisingly straightforward link between the two algorithms -- for all practical purposes, SSS is a special case of AB. Subsequent empirical evidence proves the prevailing opinion on SSS to be wrong: it is not a complicated algorithm, it does not need too much memory, and it is also not more efficient than depth-first search.
comment: PhD thesis Aske Plaat 20 June 1996. AlphaBeta, SSS*, MTD(f)
☆ Fostc3net:A Lightweight YOLOv5 Based On the Network Structure Optimization
Transmission line detection technology is crucial for automatic monitoring and ensuring the safety of electrical facilities. The YOLOv5 series is currently one of the most advanced and widely used methods for object detection. However, it faces inherent challenges, such as high computational load on devices and insufficient detection accuracy. To address these concerns, this paper presents an enhanced lightweight YOLOv5 technique customized for mobile devices, specifically intended for identifying objects associated with transmission lines. The C3Ghost module is integrated into the convolutional network of YOLOv5 to reduce floating point operations per second (FLOPs) in the feature channel fusion process and improve feature expression performance. In addition, a FasterNet module is introduced to replace the c3 module in the YOLOv5 Backbone. The FasterNet module uses Partial Convolutions to process only a portion of the input channels, improving feature extraction efficiency and reducing computational overhead. To address the imbalance between simple and challenging samples in the dataset and the diversity of aspect ratios of bounding boxes, the wIoU v3 LOSS is adopted as the loss function. To validate the performance of the proposed approach, Experiments are conducted on a custom dataset of transmission line poles. The results show that the proposed model achieves a 1% increase in detection accuracy, a 13% reduction in FLOPs, and a 26% decrease in model parameters compared to the existing YOLOv5.In the ablation experiment, it was also discovered that while the Fastnet module and the CSghost module improved the precision of the original YOLOv5 baseline model, they caused a decrease in the mAP@.5-.95 metric. However, the improvement of the wIoUv3 loss function significantly mitigated the decline of the mAP@.5-.95 metric.
☆ SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning ICLR 2024
Generalized Category Discovery (GCD) aims to classify unlabelled images from both `seen' and `unseen' classes by transferring knowledge from a set of labelled `seen' class images. A key theme in existing GCD approaches is adapting large-scale pre-trained models for the GCD task. An alternate perspective, however, is to adapt the data representation itself for better alignment with the pre-trained model. As such, in this paper, we introduce a two-stage adaptation approach termed SPTNet, which iteratively optimizes model parameters (i.e., model-finetuning) and data parameters (i.e., prompt learning). Furthermore, we propose a novel spatial prompt tuning method (SPT) which considers the spatial property of image data, enabling the method to better focus on object parts, which can transfer between seen and unseen classes. We thoroughly evaluate our SPTNet on standard benchmarks and demonstrate that our method outperforms existing GCD methods. Notably, we find our method achieves an average accuracy of 61.4% on the SSB, surpassing prior state-of-the-art methods by approximately 10%. The improvement is particularly remarkable as our method yields extra parameters amounting to only 0.117% of those in the backbone architecture. Project page: https://visual-ai.github.io/sptnet.
comment: Accepted as a conference paper at ICLR 2024; Project page: https://visual-ai.github.io/sptnet
☆ Threats, Attacks, and Defenses in Machine Unlearning: A Survey
Recently, Machine Unlearning (MU) has gained considerable attention for its potential to improve AI safety by removing the influence of specific data from trained Machine Learning (ML) models. This process, known as knowledge removal, addresses concerns about data such as sensitivity, copyright restrictions, obsolescence, or low quality. This capability is also crucial for ensuring compliance with privacy regulations such as the Right To Be Forgotten (RTBF). Therefore, strategic knowledge removal mitigates the risk of harmful outcomes, safeguarding against biases, misinformation, and unauthorized data exploitation, thereby enhancing the ethical use and reliability of AI systems. Efforts have been made to design efficient unlearning approaches, with MU services being examined for integration with existing machine learning as a service (MLaaS), allowing users to submit requests to erase data. However, recent research highlights vulnerabilities in machine unlearning systems, such as information leakage and malicious unlearning requests, that can lead to significant security and privacy concerns. Moreover, extensive research indicates that unlearning methods and prevalent attacks fulfill diverse roles within MU systems. For instance, unlearning can act as a mechanism to recover models from backdoor attacks, while backdoor attacks themselves can serve as an evaluation metric for unlearning effectiveness. This underscores the intricate relationship and complex interplay between these elements in maintaining system functionality and safety. Therefore, this survey seeks to bridge the gap between the extensive number of studies on threats, attacks, and defenses in machine unlearning and the absence of a comprehensive review that categorizes their taxonomy, methods, and solutions, thus offering valuable insights for future research directions and practical implementations.
☆ PARAMANU-AYN: An Efficient Novel Generative and Instruction-tuned Language Model for Indian Legal Case Documents
In this paper, we present PARAMANU-AYN, a language model based exclusively on case documents of the Supreme Court of India, the Constitution of India, and the Indian Penal Code. The novel Auto Regressive (AR) decoder based model is pretrained from scratch at a context size of 8192. We evaluated our pretrained legal model on perplexity metrics. We also instruction-tuned our pretrained model on a set of 10,763 instructions covering various legal tasks such as legal reasoning, judgement explanation, legal clause generation, legal drafting, legal contract drafting, case summarization, constitutional question-answering, etc. We also evaluated the responses of prompts for instruction-tuned models by GPT-3.5-Turbo on clarity, relevance, completeness, and legal reasoning metrics in a scale of 10. Our model can be run on CPU and achieved 42.46 tokens/sec CPU inference speed. We found that our models, despite not being pretrained on legal books, various legal contracts, and legal documents, were able to learn the domain knowledge required for drafting various legal contracts and legal clauses, and generalize to draft legal contracts and legal clauses with limited instruction tuning. Hence, we conclude that for a strong domain-specialized generative language model (such as legal), very large amounts of data are not required to develop models from scratch. We believe that this work is the first attempt to make a dedicated generative legal language model from scratch for Indian Supreme Court jurisdiction or in legal NLP overall. We plan to release our Paramanu-Ayn model at https://www.bharatgpts.com.
☆ Learning User Embeddings from Human Gaze for Personalised Saliency Prediction
Reusable embeddings of user behaviour have shown significant performance improvements for the personalised saliency prediction task. However, prior works require explicit user characteristics and preferences as input, which are often difficult to obtain. We present a novel method to extract user embeddings from pairs of natural images and corresponding saliency maps generated from a small amount of user-specific eye tracking data. At the core of our method is a Siamese convolutional neural encoder that learns the user embeddings by contrasting the image and personal saliency map pairs of different users. Evaluations on two public saliency datasets show that the generated embeddings have high discriminative power, are effective at refining universal saliency maps to the individual users, and generalise well across users and images. Finally, based on our model's ability to encode individual user characteristics, our work points towards other applications that can benefit from reusable embeddings of gaze behaviour.
☆ Dynamic Resource Allocation for Virtual Machine Migration Optimization using Machine Learning
The paragraph is grammatically correct and logically coherent. It discusses the importance of mobile terminal cloud computing migration technology in meeting the demands of evolving computer and cloud computing technologies. It emphasizes the need for efficient data access and storage, as well as the utilization of cloud computing migration technology to prevent additional time delays. The paragraph also highlights the contributions of cloud computing migration technology to expanding cloud computing services. Additionally, it acknowledges the role of virtualization as a fundamental capability of cloud computing while emphasizing that cloud computing and virtualization are not inherently interconnected. Finally, it introduces machine learning-based virtual machine migration optimization and dynamic resource allocation as a critical research direction in cloud computing, citing the limitations of static rules or manual settings in traditional cloud computing environments. Overall, the paragraph effectively communicates the importance of machine learning technology in addressing resource allocation and virtual machine migration challenges in cloud computing.
☆ No more optimization rules: LLM-enabled policy-based multi-modal query optimizer (version 1)
Large language model (LLM) has marked a pivotal moment in the field of machine learning and deep learning. Recently its capability for query planning has been investigated, including both single-modal and multi-modal queries. However, there is no work on the query optimization capability of LLM. As a critical (or could even be the most important) step that significantly impacts the execution performance of the query plan, such analysis and attempts should not be missed. From another aspect, existing query optimizers are usually rule-based or rule-based + cost-based, i.e., they are dependent on manually created rules to complete the query plan rewrite/transformation. Given the fact that modern optimizers include hundreds to thousands of rules, designing a multi-modal query optimizer following a similar way is significantly time-consuming since we will have to enumerate as many multi-modal optimization rules as possible, which has not been well addressed today. In this paper, we investigate the query optimization ability of LLM and use LLM to design LaPuda, a novel LLM and Policy based multi-modal query optimizer. Instead of enumerating specific and detailed rules, LaPuda only needs a few abstract policies to guide LLM in the optimization, by which much time and human effort are saved. Furthermore, to prevent LLM from making mistakes or negative optimization, we borrow the idea of gradient descent and propose a guided cost descent (GCD) algorithm to perform the optimization, such that the optimization can be kept in the correct direction. In our evaluation, our methods consistently outperform the baselines in most cases. For example, the optimized plans generated by our methods result in 1~3x higher execution speed than those by the baselines.
comment: Yifan and Haodi contribute equally to the work
☆ A Large Language Model Enhanced Sequential Recommender for Joint Video and Comment Recommendation
In online video platforms, reading or writing comments on interesting videos has become an essential part of the video watching experience. However, existing video recommender systems mainly model users' interaction behaviors with videos, lacking consideration of comments in user behavior modeling. In this paper, we propose a novel recommendation approach called LSVCR by leveraging user interaction histories with both videos and comments, so as to jointly conduct personalized video and comment recommendation. Specifically, our approach consists of two key components, namely sequential recommendation (SR) model and supplemental large language model (LLM) recommender. The SR model serves as the primary recommendation backbone (retained in deployment) of our approach, allowing for efficient user preference modeling. Meanwhile, we leverage the LLM recommender as a supplemental component (discarded in deployment) to better capture underlying user preferences from heterogeneous interaction behaviors. In order to integrate the merits of the SR model and the supplemental LLM recommender, we design a twostage training paradigm. The first stage is personalized preference alignment, which aims to align the preference representations from both components, thereby enhancing the semantics of the SR model. The second stage is recommendation-oriented fine-tuning, in which the alignment-enhanced SR model is fine-tuned according to specific objectives. Extensive experiments in both video and comment recommendation tasks demonstrate the effectiveness of LSVCR. Additionally, online A/B testing on the KuaiShou platform verifies the actual benefits brought by our approach. In particular, we achieve a significant overall gain of 4.13% in comment watch time.
☆ Find n' Propagate: Open-Vocabulary 3D Object Detection in Urban Environments
In this work, we tackle the limitations of current LiDAR-based 3D object detection systems, which are hindered by a restricted class vocabulary and the high costs associated with annotating new object classes. Our exploration of open-vocabulary (OV) learning in urban environments aims to capture novel instances using pre-trained vision-language models (VLMs) with multi-sensor data. We design and benchmark a set of four potential solutions as baselines, categorizing them into either top-down or bottom-up approaches based on their input data strategies. While effective, these methods exhibit certain limitations, such as missing novel objects in 3D box estimation or applying rigorous priors, leading to biases towards objects near the camera or of rectangular geometries. To overcome these limitations, we introduce a universal \textsc{Find n' Propagate} approach for 3D OV tasks, aimed at maximizing the recall of novel objects and propagating this detection capability to more distant areas thereby progressively capturing more. In particular, we utilize a greedy box seeker to search against 3D novel boxes of varying orientations and depth in each generated frustum and ensure the reliability of newly identified boxes by cross alignment and density ranker. Additionally, the inherent bias towards camera-proximal objects is alleviated by the proposed remote simulator, which randomly diversifies pseudo-labeled novel instances in the self-training process, combined with the fusion of base samples in the memory bank. Extensive experiments demonstrate a 53% improvement in novel recall across diverse OV settings, VLMs, and 3D detectors. Notably, we achieve up to a 3.97-fold increase in Average Precision (AP) for novel object classes. The source code is made available in the supplementary material.
☆ VCounselor: A Psychological Intervention Chat Agent Based on a Knowledge-Enhanced Large Language Model
Conversational artificial intelligence can already independently engage in brief conversations with clients with psychological problems and provide evidence-based psychological interventions. The main objective of this study is to improve the effectiveness and credibility of the large language model in psychological intervention by creating a specialized agent, the VCounselor, to address the limitations observed in popular large language models such as ChatGPT in domain applications. We achieved this goal by proposing a new affective interaction structure and knowledge-enhancement structure. In order to evaluate VCounselor, this study compared the general large language model, the fine-tuned large language model, and VCounselor's knowledge-enhanced large language model. At the same time, the general large language model and the fine-tuned large language model will also be provided with an avatar to compare them as an agent with VCounselor. The comparison results indicated that the affective interaction structure and knowledge-enhancement structure of VCounselor significantly improved the effectiveness and credibility of the psychological intervention, and VCounselor significantly provided positive tendencies for clients' emotions. The conclusion of this study strongly supports that VConselor has a significant advantage in providing psychological support to clients by being able to analyze the patient's problems with relative accuracy and provide professional-level advice that enhances support for clients.
comment: 24 pages, 6 figures
☆ What explains the success of cross-modal fine-tuning with ORCA?
ORCA (Shen et al., 2023) is a recent technique for cross-modal fine-tuning, i.e., applying pre-trained transformer models to modalities beyond their training data. The technique consists primarily of training an embedder and fine-tuning the embedder and model. Despite its high performance on a variety of downstream tasks, we do not understand precisely how each of these components contribute to ORCA's success. Therefore, we run a series of ablations and find that embedder training does not help 2D tasks at all, contrary to what the original paper posits. In 1D tasks, some amount of embedder training is necessary but more is not better. In 4 out of 6 datasets we experiment with, it is model fine-tuning that makes the biggest difference. Through our ablations and baselines, we contribute a better understanding of the individual components of ORCA.
☆ Compress3D: a Compressed Latent Space for 3D Generation from a Single Image
3D generation has witnessed significant advancements, yet efficiently producing high-quality 3D assets from a single image remains challenging. In this paper, we present a triplane autoencoder, which encodes 3D models into a compact triplane latent space to effectively compress both the 3D geometry and texture information. Within the autoencoder framework, we introduce a 3D-aware cross-attention mechanism, which utilizes low-resolution latent representations to query features from a high-resolution 3D feature volume, thereby enhancing the representation capacity of the latent space. Subsequently, we train a diffusion model on this refined latent space. In contrast to solely relying on image embedding for 3D generation, our proposed method advocates for the simultaneous utilization of both image embedding and shape embedding as conditions. Specifically, the shape embedding is estimated via a diffusion prior model conditioned on the image embedding. Through comprehensive experiments, we demonstrate that our method outperforms state-of-the-art algorithms, achieving superior performance while requiring less training data and time. Our approach enables the generation of high-quality 3D assets in merely 7 seconds on a single A100 GPU.
☆ Have You Poisoned My Data? Defending Neural Networks against Data Poisoning
The unprecedented availability of training data fueled the rapid development of powerful neural networks in recent years. However, the need for such large amounts of data leads to potential threats such as poisoning attacks: adversarial manipulations of the training data aimed at compromising the learned model to achieve a given adversarial goal. This paper investigates defenses against clean-label poisoning attacks and proposes a novel approach to detect and filter poisoned datapoints in the transfer learning setting. We define a new characteristic vector representation of datapoints and show that it effectively captures the intrinsic properties of the data distribution. Through experimental analysis, we demonstrate that effective poisons can be successfully differentiated from clean points in the characteristic vector space. We thoroughly evaluate our proposed approach and compare it to existing state-of-the-art defenses using multiple architectures, datasets, and poison budgets. Our evaluation shows that our proposal outperforms existing approaches in defense rate and final trained model performance across all experimental settings.
comment: Paper accepted for publication at European Symposium on Research in Computer Security (ESORICS) 2024
☆ Motion Generation from Fine-grained Textual Descriptions
The task of text2motion is to generate motion sequences from given textual descriptions, where a model should explore the interactions between natural language instructions and human body movements. While most existing works are confined to coarse-grained motion descriptions (e.g., "A man squats."), fine-grained ones specifying movements of relevant body parts are barely explored. Models trained with coarse texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure in generating motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset with fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with delicate prompts. Accordingly, we design a new text2motion model, FineMotionDiffuse, which makes full use of fine-grained textual information. Our experiments show that FineMotionDiffuse trained on FineHumanML3D acquires good results in quantitative evaluation. We also find this model can better generate spatially/chronologically composite motions by learning the implicit mappings from simple descriptions to the corresponding basic motions.
☆ What if...?: Counterfactual Inception to Mitigate Hallucination Effects in Large Multimodal Models
This paper presents a way of enhancing the reliability of Large Multimodal Models (LMMs) in addressing hallucination effects, where models generate incorrect or unrelated responses. Without additional instruction tuning paradigm, we introduce Counterfactual Inception, a novel method that implants counterfactual thoughts into LMMs using carefully chosen, misaligned counterfactual keywords. This method is grounded in the concept of counterfactual thinking, a cognitive process where humans consider alternative realities and outcomes. By applying this human-like reasoning mechanism to LMMs, we aim to reduce hallucination effects and improve the models' trustworthiness. We also propose Dual-modality Verification Process (DVP), a rigorous framework for selecting optimal counterfactual keywords to trigger counterfactual thinking into LMMs, concurrently considering visual and linguistic context. Our extensive experiments across various LMMs, including both open-source and proprietary models, corroborate that our method significantly mitigates hallucination phenomena across different datasets.
comment: under review, code available: https://github.com/IVY-LVLM/Counterfactual-Inception
☆ Scale Decoupled Distillation CVPR2024
Logit knowledge distillation attracts increasing attention due to its practicality in recent studies. However, it often suffers inferior performance compared to the feature knowledge distillation. In this paper, we argue that existing logit-based methods may be sub-optimal since they only leverage the global logit output that couples multiple semantic knowledge. This may transfer ambiguous knowledge to the student and mislead its learning. To this end, we propose a simple but effective method, i.e., Scale Decoupled Distillation (SDD), for logit knowledge distillation. SDD decouples the global logit output into multiple local logit outputs and establishes distillation pipelines for them. This helps the student to mine and inherit fine-grained and unambiguous logit knowledge. Moreover, the decoupled knowledge can be further divided into consistent and complementary logit knowledge that transfers the semantic information and sample ambiguity, respectively. By increasing the weight of complementary parts, SDD can guide the student to focus more on ambiguous samples, improving its discrimination ability. Extensive experiments on several benchmark datasets demonstrate the effectiveness of SDD for wide teacher-student pairs, especially in the fine-grained classification task. Code is available at: https://github.com/shicaiwei123/SDD-CVPR2024
comment: Accepted to CVPR2024 10 pages 6figure
☆ VSTAR: Generative Temporal Nursing for Longer Dynamic Video Synthesis
Despite tremendous progress in the field of text-to-video (T2V) synthesis, open-sourced T2V diffusion models struggle to generate longer videos with dynamically varying and evolving content. They tend to synthesize quasi-static videos, ignoring the necessary visual change-over-time implied in the text prompt. At the same time, scaling these models to enable longer, more dynamic video synthesis often remains computationally intractable. To address this challenge, we introduce the concept of Generative Temporal Nursing (GTN), where we aim to alter the generative process on the fly during inference to improve control over the temporal dynamics and enable generation of longer videos. We propose a method for GTN, dubbed VSTAR, which consists of two key ingredients: 1) Video Synopsis Prompting (VSP) - automatic generation of a video synopsis based on the original single prompt leveraging LLMs, which gives accurate textual guidance to different visual states of longer videos, and 2) Temporal Attention Regularization (TAR) - a regularization technique to refine the temporal attention units of the pre-trained T2V diffusion models, which enables control over the video dynamics. We experimentally showcase the superiority of the proposed approach in generating longer, visually appealing videos over existing open-sourced T2V models. We additionally analyze the temporal attention maps realized with and without VSTAR, demonstrating the importance of applying our method to mitigate neglect of the desired visual change over time.
comment: Project page: https://yumengli007.github.io/VSTAR
☆ Deepfake Detection without Deepfakes: Generalization via Synthetic Frequency Patterns Injection
Deepfake detectors are typically trained on large sets of pristine and generated images, resulting in limited generalization capacity; they excel at identifying deepfakes created through methods encountered during training but struggle with those generated by unknown techniques. This paper introduces a learning approach aimed at significantly enhancing the generalization capabilities of deepfake detectors. Our method takes inspiration from the unique "fingerprints" that image generation processes consistently introduce into the frequency domain. These fingerprints manifest as structured and distinctly recognizable frequency patterns. We propose to train detectors using only pristine images injecting in part of them crafted frequency patterns, simulating the effects of various deepfake generation techniques without being specific to any. These synthetic patterns are based on generic shapes, grids, or auras. We evaluated our approach using diverse architectures across 25 different generation methods. The models trained with our approach were able to perform state-of-the-art deepfake detection, demonstrating also superior generalization capabilities in comparison with previous methods. Indeed, they are untied to any specific generation technique and can effectively identify deepfakes regardless of how they were made.
☆ HyperLLaVA: Dynamic Visual and Language Expert Tuning for Multimodal Large Language Models
Recent advancements indicate that scaling up Multimodal Large Language Models (MLLMs) effectively enhances performance on downstream multimodal tasks. The prevailing MLLM paradigm, \emph{e.g.}, LLaVA, transforms visual features into text-like tokens using a \emph{static} vision-language mapper, thereby enabling \emph{static} LLMs to develop the capability to comprehend visual information through visual instruction tuning. Although promising, the \emph{static} tuning strategy~\footnote{The static tuning refers to the trained model with static parameters.} that shares the same parameters may constrain performance across different downstream multimodal tasks. In light of this, we introduce HyperLLaVA, which involves adaptive tuning of the projector and LLM parameters, in conjunction with a dynamic visual expert and language expert, respectively. These experts are derived from HyperNetworks, which generates adaptive parameter shifts through visual and language guidance, enabling dynamic projector and LLM modeling in two-stage training. Our experiments demonstrate that our solution significantly surpasses LLaVA on existing MLLM benchmarks, including MME, MMBench, SEED-Bench, and LLaVA-Bench. ~\footnote{Our project is available on the link https://github.com/DCDmllm/HyperLLaVA}.
☆ Robustness Verifcation in Neural Networks
In this paper we investigate formal verification problems for Neural Network computations. Of central importance will be various robustness and minimization problems such as: Given symbolic specifications of allowed inputs and outputs in form of Linear Programming instances, one question is whether there do exist valid inputs such that the network computes a valid output? And does this property hold for all valid inputs? Do two given networks compute the same function? Is there a smaller network computing the same function? The complexity of these questions have been investigated recently from a practical point of view and approximated by heuristic algorithms. We complement these achievements by giving a theoretical framework that enables us to interchange security and efficiency questions in neural networks and analyze their computational complexities. We show that the problems are conquerable in a semi-linear setting, meaning that for piecewise linear activation functions and when the sum- or maximum metric is used, most of them are in P or in NP at most.
comment: 16 pages, 1 figure
☆ Agent Group Chat: An Interactive Group Chat Simulacra For Better Eliciting Collective Emergent Behavior
To investigate the role of language in human collective behaviors, we developed the Agent Group Chat simulation to simulate linguistic interactions among multi-agent in different settings. Agents are asked to free chat in this simulation for their own purposes based on their character setting, aiming to see agents exhibit emergent behaviours that are both unforeseen and significant. Four narrative scenarios, Inheritance Disputes, Law Court Debates, Philosophical Discourses, Movie Casting Contention, are integrated into Agent Group Chat to evaluate its support for diverse storylines. By configuring specific environmental settings within Agent Group Chat, we are able to assess whether agents exhibit behaviors that align with human expectations. We evaluate the disorder within the environment by computing the n-gram Shannon entropy of all the content speak by characters. Our findings reveal that under the premise of agents possessing substantial alignment with human expectations, facilitating more extensive information exchange within the simulation ensures greater orderliness amidst diversity, which leads to the emergence of more unexpected and meaningful emergent behaviors. The code is open source in https://github.com/MikeGu721/AgentGroup, and online platform will be open soon.
☆ Caching-Augmented Lifelong Multi-Agent Path Finding
Multi-Agent Path Finding (MAPF), which involves finding collision-free paths for multiple robots, is crucial in various applications. Lifelong MAPF, where targets are reassigned to agents as soon as they complete their initial objectives, offers a more accurate approximation of real-world warehouse planning. In this paper, we present a novel mechanism named Caching-Augmented Lifelong MAPF (CAL-MAPF), designed to improve the performance of Lifelong MAPF. We have developed a new map grid type called cache for temporary item storage and replacement and designed a lock mechanism for it to improve the stability of the planning solution. This cache mechanism was evaluated using various cache replacement policies and a spectrum of input task distributions. We identified three main factors significantly impacting CAL-MAPF performance through experimentation: suitable input task distribution, high cache hit rate, and smooth traffic. Overall, CAL-MAPF has demonstrated potential for performance improvements in certain task distributions, maps and agent configurations.
☆ S2DM: Sector-Shaped Diffusion Models for Video Generation
Diffusion models have achieved great success in image generation. However, when leveraging this idea for video generation, we face significant challenges in maintaining the consistency and continuity across video frames. This is mainly caused by the lack of an effective framework to align frames of videos with desired temporal features while preserving consistent semantic and stochastic features. In this work, we propose a novel Sector-Shaped Diffusion Model (S2DM) whose sector-shaped diffusion region is formed by a set of ray-shaped reverse diffusion processes starting at the same noise point. S2DM can generate a group of intrinsically related data sharing the same semantic and stochastic features while varying on temporal features with appropriate guided conditions. We apply S2DM to video generation tasks, and explore the use of optical flow as temporal conditions. Our experimental results show that S2DM outperforms many existing methods in the task of video generation without any temporal-feature modelling modules. For text-to-video generation tasks where temporal conditions are not explicitly given, we propose a two-stage generation strategy which can decouple the generation of temporal features from semantic-content features. We show that, without additional training, our model integrated with another temporal conditions generative model can still achieve comparable performance with existing works. Our results can be viewd at https://s2dm.github.io/S2DM/.
comment: 17 pages, 6 figures
☆ DOR3D-Net: Dense Ordinal Regression Network for 3D Hand Pose Estimation
Depth-based 3D hand pose estimation is an important but challenging research task in human-machine interaction community. Recently, dense regression methods have attracted increasing attention in 3D hand pose estimation task, which provide a low computational burden and high accuracy regression way by densely regressing hand joint offset maps. However, large-scale regression offset values are often affected by noise and outliers, leading to a significant drop in accuracy. To tackle this, we re-formulate 3D hand pose estimation as a dense ordinal regression problem and propose a novel Dense Ordinal Regression 3D Pose Network (DOR3D-Net). Specifically, we first decompose offset value regression into sub-tasks of binary classifications with ordinal constraints. Then, each binary classifier can predict the probability of a binary spatial relationship relative to joint, which is easier to train and yield much lower level of noise. The estimated hand joint positions are inferred by aggregating the ordinal regression results at local positions with a weighted sum. Furthermore, both joint regression loss and ordinal regression loss are used to train our DOR3D-Net in an end-to-end manner. Extensive experiments on public datasets (ICVL, MSRA, NYU and HANDS2017) show that our design provides significant improvements over SOTA methods.
☆ Byzantine-resilient Federated Learning With Adaptivity to Data Heterogeneity
This paper deals with federated learning (FL) in the presence of malicious Byzantine attacks and data heterogeneity. A novel Robust Average Gradient Algorithm (RAGA) is proposed, which leverages the geometric median for aggregation and can freely select the round number for local updating. Different from most existing resilient approaches, which perform convergence analysis based on strongly-convex loss function or homogeneously distributed dataset, we conduct convergence analysis for not only strongly-convex but also non-convex loss function over heterogeneous dataset. According to our theoretical analysis, as long as the fraction of dataset from malicious users is less than half, RAGA can achieve convergence at rate $\mathcal{O}({1}/{T^{2/3- \delta}})$ where $T$ is the iteration number and $\delta \in (0, 2/3)$ for non-convex loss function, and at linear rate for strongly-convex loss function. Moreover, stationary point or global optimal solution is proved to obtainable as data heterogeneity vanishes. Experimental results corroborate the robustness of RAGA to Byzantine attacks and verifies the advantage of RAGA over baselines on convergence performance under various intensity of Byzantine attacks, for heterogeneous dataset.
☆ LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models
Efficient fine-tuning is vital for adapting large language models (LLMs) to downstream tasks. However, it requires non-trivial efforts to implement these methods on different models. We present LlamaFactory, a unified framework that integrates a suite of cutting-edge efficient training methods. It allows users to flexibly customize the fine-tuning of 100+ LLMs without the need for coding through the built-in web UI LlamaBoard. We empirically validate the efficiency and effectiveness of our framework on language modeling and text generation tasks. It has been released at https://github.com/hiyouga/LLaMA-Factory and already received over 13,000 stars and 1,600 forks.
comment: 12 pages, preprint
☆ Clinical information extraction for Low-resource languages with Few-shot learning using Pre-trained language models and Prompting
Automatic extraction of medical information from clinical documents poses several challenges: high costs of required clinical expertise, limited interpretability of model predictions, restricted computational resources and privacy regulations. Recent advances in domain-adaptation and prompting methods showed promising results with minimal training data using lightweight masked language models, which are suited for well-established interpretability methods. We are first to present a systematic evaluation of these methods in a low-resource setting, by performing multi-class section classification on German doctor's letters. We conduct extensive class-wise evaluations supported by Shapley values, to validate the quality of our small training data set and to ensure the interpretability of model predictions. We demonstrate that a lightweight, domain-adapted pretrained model, prompted with just 20 shots, outperforms a traditional classification model by 30.5% accuracy. Our results serve as a process-oriented guideline for clinical information extraction projects working with low-resource.
☆ Computational Models to Study Language Processing in the Human Brain: A Survey
Despite differing from the human language processing mechanism in implementation and algorithms, current language models demonstrate remarkable human-like or surpassing language capabilities. Should computational language models be employed in studying the brain, and if so, when and how? To delve into this topic, this paper reviews efforts in using computational models for brain research, highlighting emerging trends. To ensure a fair comparison, the paper evaluates various computational models using consistent metrics on the same dataset. Our analysis reveals that no single model outperforms others on all datasets, underscoring the need for rich testing datasets and rigid experimental control to draw robust conclusions in studies involving computational models.
☆ Incentivizing News Consumption on Social Media Platforms Using Large Language Models and Realistic Bot Accounts
Polarization, declining trust, and wavering support for democratic norms are pressing threats to U.S. democracy. Exposure to verified and quality news may lower individual susceptibility to these threats and make citizens more resilient to misinformation, populism, and hyperpartisan rhetoric. This project examines how to enhance users' exposure to and engagement with verified and ideologically balanced news in an ecologically valid setting. We rely on a large-scale two-week long field experiment (from 1/19/2023 to 2/3/2023) on 28,457 Twitter users. We created 28 bots utilizing GPT-2 that replied to users tweeting about sports, entertainment, or lifestyle with a contextual reply containing two hardcoded elements: a URL to the topic-relevant section of quality news organization and an encouragement to follow its Twitter account. To further test differential effects by gender of the bots, treated users were randomly assigned to receive responses by bots presented as female or male. We examine whether our over-time intervention enhances the following of news media organization, the sharing and the liking of news content and the tweeting about politics and the liking of political content. We find that the treated users followed more news accounts and the users in the female bot treatment were more likely to like news content than the control. Most of these results, however, were small in magnitude and confined to the already politically interested Twitter users, as indicated by their pre-treatment tweeting about politics. These findings have implications for social media and news organizations, and also offer direction for future work on how Large Language Models and other computational interventions can effectively enhance individual on-platform engagement with quality news and public affairs.
BadEdit: Backdooring large language models by model editing ICLR 2024
Mainstream backdoor attack methods typically demand substantial tuning data for poisoning, limiting their practicality and potentially degrading the overall performance when applied to Large Language Models (LLMs). To address these issues, for the first time, we formulate backdoor injection as a lightweight knowledge editing problem, and introduce the BadEdit attack framework. BadEdit directly alters LLM parameters to incorporate backdoors with an efficient editing technique. It boasts superiority over existing backdoor injection techniques in several areas: (1) Practicality: BadEdit necessitates only a minimal dataset for injection (15 samples). (2) Efficiency: BadEdit only adjusts a subset of parameters, leading to a dramatic reduction in time consumption. (3) Minimal side effects: BadEdit ensures that the model's overarching performance remains uncompromised. (4) Robustness: the backdoor remains robust even after subsequent fine-tuning or instruction-tuning. Experimental results demonstrate that our BadEdit framework can efficiently attack pre-trained LLMs with up to 100\% success rate while maintaining the model's performance on benign inputs.
comment: ICLR 2024
☆ USE: Dynamic User Modeling with Stateful Sequence Models
User embeddings play a crucial role in user engagement forecasting and personalized services. Recent advances in sequence modeling have sparked interest in learning user embeddings from behavioral data. Yet behavior-based user embedding learning faces the unique challenge of dynamic user modeling. As users continuously interact with the apps, user embeddings should be periodically updated to account for users' recent and long-term behavior patterns. Existing methods highly rely on stateless sequence models that lack memory of historical behavior. They have to either discard historical data and use only the most recent data or reprocess the old and new data jointly. Both cases incur substantial computational overhead. To address this limitation, we introduce User Stateful Embedding (USE). USE generates user embeddings and reflects users' evolving behaviors without the need for exhaustive reprocessing by storing previous model states and revisiting them in the future. Furthermore, we introduce a novel training objective named future W-behavior prediction to transcend the limitations of next-token prediction by forecasting a broader horizon of upcoming user behaviors. By combining it with the Same User Prediction, a contrastive learning-based objective that predicts whether different segments of behavior sequences belong to the same user, we further improve the embeddings' distinctiveness and representativeness. We conducted experiments on 8 downstream tasks using Snapchat users' behavioral logs in both static (i.e., fixed user behavior sequences) and dynamic (i.e., periodically updated user behavior sequences) settings. We demonstrate USE's superior performance over established baselines. The results underscore USE's effectiveness and efficiency in integrating historical and recent user behavior sequences into user embeddings in dynamic user modeling.
☆ FissionFusion: Fast Geometric Generation and Hierarchical Souping for Medical Image Analysis
The scarcity of well-annotated medical datasets requires leveraging transfer learning from broader datasets like ImageNet or pre-trained models like CLIP. Model soups averages multiple fine-tuned models aiming to improve performance on In-Domain (ID) tasks and enhance robustness against Out-of-Distribution (OOD) datasets. However, applying these methods to the medical imaging domain faces challenges and results in suboptimal performance. This is primarily due to differences in error surface characteristics that stem from data complexities such as heterogeneity, domain shift, class imbalance, and distributional shifts between training and testing phases. To address this issue, we propose a hierarchical merging approach that involves local and global aggregation of models at various levels based on models' hyperparameter configurations. Furthermore, to alleviate the need for training a large number of models in the hyperparameter search, we introduce a computationally efficient method using a cyclical learning rate scheduler to produce multiple models for aggregation in the weight space. Our method demonstrates significant improvements over the model souping approach across multiple datasets (around 6% gain in HAM10000 and CheXpert datasets) while maintaining low computational costs for model generation and selection. Moreover, we achieve better results on OOD datasets than model soups. The code is available at https://github.com/BioMedIA-MBZUAI/FissionFusion.
☆ Learning Novel View Synthesis from Heterogeneous Low-light Captures
Neural radiance field has achieved fundamental success in novel view synthesis from input views with the same brightness level captured under fixed normal lighting. Unfortunately, synthesizing novel views remains to be a challenge for input views with heterogeneous brightness level captured under low-light condition. The condition is pretty common in the real world. It causes low-contrast images where details are concealed in the darkness and camera sensor noise significantly degrades the image quality. To tackle this problem, we propose to learn to decompose illumination, reflectance, and noise from input views according to that reflectance remains invariant across heterogeneous views. To cope with heterogeneous brightness and noise levels across multi-views, we learn an illumination embedding and optimize a noise map individually for each view. To allow intuitive editing of the illumination, we design an illumination adjustment module to enable either brightening or darkening of the illumination component. Comprehensive experiments demonstrate that this approach enables effective intrinsic decomposition for low-light multi-view noisy images and achieves superior visual quality and numerical performance for synthesizing novel views compared to state-of-the-art methods.
☆ Adaptive Ensembles of Fine-Tuned Transformers for LLM-Generated Text Detection
Large language models (LLMs) have reached human-like proficiency in generating diverse textual content, underscoring the necessity for effective fake text detection to avoid potential risks such as fake news in social media. Previous research has mostly tested single models on in-distribution datasets, limiting our understanding of how these models perform on different types of data for LLM-generated text detection task. We researched this by testing five specialized transformer-based models on both in-distribution and out-of-distribution datasets to better assess their performance and generalizability. Our results revealed that single transformer-based classifiers achieved decent performance on in-distribution dataset but limited generalization ability on out-of-distribution dataset. To improve it, we combined the individual classifiers models using adaptive ensemble algorithms, which improved the average accuracy significantly from 91.8% to 99.2% on an in-distribution test set and from 62.9% to 72.5% on an out-of-distribution test set. The results indicate the effectiveness, good generalization ability, and great potential of adaptive ensemble algorithms in LLM-generated text detection.
☆ Hyacinth6B: A large language model for Traditional Chinese
This research's primary motivation of this study is to address the high hardware and computational demands typically associated with LLMs.Therefore,our goal is to find a balance between model lightness and performance,striving to maximize performance while using a comparatively lightweight model. Hyacinth6B was developed with this objective in mind,aiming to fully leverage the core capabilities of LLMs without incurring substantial resource costs, effectively pushing the boundaries of smaller model's performance. The training approach involves parameter efficient finetuning using the LoRA method.
comment: 14pages
☆ Polaris: A Safety-focused LLM Constellation Architecture for Healthcare
We develop Polaris, the first safety-focused LLM constellation for real-time patient-AI healthcare conversations. Unlike prior LLM works in healthcare focusing on tasks like question answering, our work specifically focuses on long multi-turn voice conversations. Our one-trillion parameter constellation system is composed of several multibillion parameter LLMs as co-operative agents: a stateful primary agent that focuses on driving an engaging conversation and several specialist support agents focused on healthcare tasks performed by nurses to increase safety and reduce hallucinations. We develop a sophisticated training protocol for iterative co-training of the agents that optimize for diverse objectives. We train our models on proprietary data, clinical care plans, healthcare regulatory documents, medical manuals, and other medical reasoning documents. We align our models to speak like medical professionals, using organic healthcare conversations and simulated ones between patient actors and experienced nurses. This allows our system to express unique capabilities such as rapport building, trust building, empathy and bedside manner. Finally, we present the first comprehensive clinician evaluation of an LLM system for healthcare. We recruited over 1100 U.S. licensed nurses and over 130 U.S. licensed physicians to perform end-to-end conversational evaluations of our system by posing as patients and rating the system on several measures. We demonstrate Polaris performs on par with human nurses on aggregate across dimensions such as medical safety, clinical readiness, conversational quality, and bedside manner. Additionally, we conduct a challenging task-based evaluation of the individual specialist support agents, where we demonstrate our LLM agents significantly outperform a much larger general-purpose LLM (GPT-4) as well as from its own medium-size class (LLaMA-2 70B).
☆ Multi-Robot Connected Fermat Spiral Coverage ICAPS24
We introduce the Multi-Robot Connected Fermat Spiral (MCFS), a novel algorithmic framework for Multi-Robot Coverage Path Planning (MCPP) that adapts Connected Fermat Spiral (CFS) from the computer graphics community to multi-robot coordination for the first time. MCFS uniquely enables the orchestration of multiple robots to generate coverage paths that contour around arbitrarily shaped obstacles, a feature that is notably lacking in traditional methods. Our framework not only enhances area coverage and optimizes task performance, particularly in terms of makespan, for workspaces rich in irregular obstacles but also addresses the challenges of path continuity and curvature critical for non-holonomic robots by generating smooth paths without decomposing the workspace. MCFS solves MCPP by constructing a graph of isolines and transforming MCPP into a combinatorial optimization problem, aiming to minimize the makespan while covering all vertices. Our contributions include developing a unified CFS version for scalable and adaptable MCPP, extending it to MCPP with novel optimization techniques for cost reduction and path continuity and smoothness, and demonstrating through extensive experiments that MCFS outperforms existing MCPP methods in makespan, path curvature, coverage ratio, and overlapping ratio. Our research marks a significant step in MCPP, showcasing the fusion of computer graphics and automated planning principles to advance the capabilities of multi-robot systems in complex environments. Our code is available at https://github.com/reso1/MCFS.
comment: accepted to ICAPS24
☆ Mapping LLM Security Landscapes: A Comprehensive Stakeholder Risk Assessment Proposal
The rapid integration of Large Language Models (LLMs) across diverse sectors has marked a transformative era, showcasing remarkable capabilities in text generation and problem-solving tasks. However, this technological advancement is accompanied by significant risks and vulnerabilities. Despite ongoing security enhancements, attackers persistently exploit these weaknesses, casting doubts on the overall trustworthiness of LLMs. Compounding the issue, organisations are deploying LLM-integrated systems without understanding the severity of potential consequences. Existing studies by OWASP and MITRE offer a general overview of threats and vulnerabilities but lack a method for directly and succinctly analysing the risks for security practitioners, developers, and key decision-makers who are working with this novel technology. To address this gap, we propose a risk assessment process using tools like the OWASP risk rating methodology which is used for traditional systems. We conduct scenario analysis to identify potential threat agents and map the dependent system components against vulnerability factors. Through this analysis, we assess the likelihood of a cyberattack. Subsequently, we conduct a thorough impact analysis to derive a comprehensive threat matrix. We also map threats against three key stakeholder groups: developers engaged in model fine-tuning, application developers utilizing third-party APIs, and end users. The proposed threat matrix provides a holistic evaluation of LLM-related risks, enabling stakeholders to make informed decisions for effective mitigation strategies. Our outlined process serves as an actionable and comprehensive tool for security practitioners, offering insights for resource management and enhancing the overall system security.
comment: 10 pages, 1 figure, 3 tables
☆ Building Optimal Neural Architectures using Interpretable Knowledge CVPR'24
Neural Architecture Search is a costly practice. The fact that a search space can span a vast number of design choices with each architecture evaluation taking nontrivial overhead makes it hard for an algorithm to sufficiently explore candidate networks. In this paper, we propose AutoBuild, a scheme which learns to align the latent embeddings of operations and architecture modules with the ground-truth performance of the architectures they appear in. By doing so, AutoBuild is capable of assigning interpretable importance scores to architecture modules, such as individual operation features and larger macro operation sequences such that high-performance neural networks can be constructed without any need for search. Through experiments performed on state-of-the-art image classification, segmentation, and Stable Diffusion models, we show that by mining a relatively small set of evaluated architectures, AutoBuild can learn to build high-quality architectures directly or help to reduce search space to focus on relevant areas, finding better architectures that outperform both the original labeled ones and ones found by search baselines. Code available at https://github.com/Ascend-Research/AutoBuild
comment: CVPR'24; 18 Pages, 18 Figures, 3 Tables
☆ AFLoRA: Adaptive Freezing of Low Rank Adaptation in Parameter Efficient Fine-Tuning of Large Models
We present a novel Parameter-Efficient Fine-Tuning (PEFT) method, dubbed as Adaptive Freezing of Low Rank Adaptation (AFLoRA). Specifically, for each pre-trained frozen weight tensor, we add a parallel path of trainable low-rank matrices, namely a down-projection and an up-projection matrix, each of which is followed by a feature transformation vector. Based on a novel freezing score, we the incrementally freeze these projection matrices during fine-tuning to reduce the computation and alleviate over-fitting. Our experimental results demonstrate that we can achieve state-of-the-art performance with an average improvement of up to $0.85\%$ as evaluated on GLUE benchmark while yeilding up to $9.5\times$ fewer average trainable parameters. While compared in terms of runtime, AFLoRA can yield up to $1.86\times$ improvement as opposed to similar PEFT alternatives. Besides the practical utility of our approach, we provide insights on the trainability requirements of LoRA paths at different modules and the freezing schedule for the different projection matrices. Code will be released.
comment: 5 pages, 5 figures
☆ Arcee's MergeKit: A Toolkit for Merging Large Language Models
The rapid expansion of the open-source language model landscape presents an opportunity to merge the competencies of these model checkpoints by combining their parameters. Advances in transfer learning, the process of fine-tuning pre-trained models for specific tasks, has resulted in the development of vast amounts of task-specific models, typically specialized in individual tasks and unable to utilize each other's strengths. Model merging facilitates the creation of multitask models without the need for additional training, offering a promising avenue for enhancing model performance and versatility. By preserving the intrinsic capabilities of the original models, model merging addresses complex challenges in AI - including the difficulties of catastrophic forgetting and multi-task learning. To support this expanding area of research, we introduce MergeKit, a comprehensive, open-source library designed to facilitate the application of model merging strategies. MergeKit offers an extensible framework to efficiently merge models on any hardware, providing utility to researchers and practitioners. To date, thousands of models have been merged by the open-source community, leading to the creation of some of the worlds most powerful open-source model checkpoints, as assessed by the Open LLM Leaderboard. The library is accessible at https://github.com/arcee-ai/MergeKit.
comment: 11 pages, 4 figures
☆ A Unified and General Framework for Continual Learning ICLR 2024
Continual Learning (CL) focuses on learning from dynamic and changing data distributions while retaining previously acquired knowledge. Various methods have been developed to address the challenge of catastrophic forgetting, including regularization-based, Bayesian-based, and memory-replay-based techniques. However, these methods lack a unified framework and common terminology for describing their approaches. This research aims to bridge this gap by introducing a comprehensive and overarching framework that encompasses and reconciles these existing methodologies. Notably, this new framework is capable of encompassing established CL approaches as special instances within a unified and general optimization objective. An intriguing finding is that despite their diverse origins, these methods share common mathematical structures. This observation highlights the compatibility of these seemingly distinct techniques, revealing their interconnectedness through a shared underlying optimization objective. Moreover, the proposed general framework introduces an innovative concept called refresh learning, specifically designed to enhance the CL performance. This novel approach draws inspiration from neuroscience, where the human brain often sheds outdated information to improve the retention of crucial knowledge and facilitate the acquisition of new information. In essence, refresh learning operates by initially unlearning current data and subsequently relearning it. It serves as a versatile plug-in that seamlessly integrates with existing CL methods, offering an adaptable and effective enhancement to the learning process. Extensive experiments on CL benchmarks and theoretical analysis demonstrate the effectiveness of the proposed refresh learning. Code is available at \url{https://github.com/joey-wang123/CL-refresh-learning}.
comment: ICLR 2024
☆ Federated reinforcement learning for robot motion planning with zero-shot generalization
This paper considers the problem of learning a control policy for robot motion planning with zero-shot generalization, i.e., no data collection and policy adaptation is needed when the learned policy is deployed in new environments. We develop a federated reinforcement learning framework that enables collaborative learning of multiple learners and a central server, i.e., the Cloud, without sharing their raw data. In each iteration, each learner uploads its local control policy and the corresponding estimated normalized arrival time to the Cloud, which then computes the global optimum among the learners and broadcasts the optimal policy to the learners. Each learner then selects between its local control policy and that from the Cloud for next iteration. The proposed framework leverages on the derived zero-shot generalization guarantees on arrival time and safety. Theoretical guarantees on almost-sure convergence, almost consensus, Pareto improvement and optimality gap are also provided. Monte Carlo simulation is conducted to evaluate the proposed framework.
☆ Instruction Multi-Constraint Molecular Generation Using a Teacher-Student Large Language Model
While various models and computational tools have been proposed for structure and property analysis of molecules, generating molecules that conform to all desired structures and properties remains a challenge. Here, we introduce a multi-constraint molecular generation large language model, TSMMG, which, akin to a student, incorporates knowledge from various small models and tools, namely, the 'teachers'. To train TSMMG, we construct a large set of text-molecule pairs by extracting molecular knowledge from these 'teachers', enabling it to generate novel molecules that conform to the descriptions through various text prompts. We experimentally show that TSMMG remarkably performs in generating molecules meeting complex, natural language-described property requirements across two-, three-, and four-constraint tasks, with an average molecular validity of over 99% and success ratio of 88.08%, 65.27%, and 61.44%, respectively. The model also exhibits adaptability through zero-shot testing, creating molecules that satisfy combinations of properties that have not been encountered. It can comprehend text inputs with various language styles, extending beyond the confines of outlined prompts, as confirmed through empirical validation. Additionally, the knowledge distillation feature of TSMMG contributes to the continuous enhancement of small models, while the innovative approach to dataset construction effectively addresses the issues of data scarcity and quality, which positions TSMMG as a promising tool in the domains of drug discovery and materials science. Code is available at https://github.com/HHW-zhou/TSMMG.
comment: 25 pages, 4 figures
☆ Safety-Aware Reinforcement Learning for Electric Vehicle Charging Station Management in Distribution Network
The increasing integration of electric vehicles (EVs) into the grid can pose a significant risk to the distribution system operation in the absence of coordination. In response to the need for effective coordination of EVs within the distribution network, this paper presents a safety-aware reinforcement learning (RL) algorithm designed to manage EV charging stations while ensuring the satisfaction of system constraints. Unlike existing methods, our proposed algorithm does not rely on explicit penalties for constraint violations, eliminating the need for penalty coefficient tuning. Furthermore, managing EV charging stations is further complicated by multiple uncertainties, notably the variability in solar energy generation and energy prices. To address this challenge, we develop an off-policy RL algorithm to efficiently utilize data to learn patterns in such uncertain environments. Our algorithm also incorporates a maximum entropy framework to enhance the RL algorithm's exploratory process, preventing convergence to local optimal solutions. Simulation results demonstrate that our algorithm outperforms traditional RL algorithms in managing EV charging in the distribution network.
comment: 2024 IEEE Power & Energy Society General Meeting (PESGM)
☆ Self-Attention Based Semantic Decomposition in Vector Symbolic Architectures
Vector Symbolic Architectures (VSAs) have emerged as a novel framework for enabling interpretable machine learning algorithms equipped with the ability to reason and explain their decision processes. The basic idea is to represent discrete information through high dimensional random vectors. Complex data structures can be built up with operations over vectors such as the "binding" operation involving element-wise vector multiplication, which associates data together. The reverse task of decomposing the associated elements is a combinatorially hard task, with an exponentially large search space. The main algorithm for performing this search is the resonator network, inspired by Hopfield network-based memory search operations. In this work, we introduce a new variant of the resonator network, based on self-attention based update rules in the iterative search problem. This update rule, based on the Hopfield network with log-sum-exp energy function and norm-bounded states, is shown to substantially improve the performance and rate of convergence. As a result, our algorithm enables a larger capacity for associative memory, enabling applications in many tasks like perception based pattern recognition, scene decomposition, and object reasoning. We substantiate our algorithm with a thorough evaluation and comparisons to baselines.
☆ Nellie: Automated organelle segmentation, tracking, and hierarchical feature extraction in 2D/3D live-cell microscopy
The analysis of dynamic organelles remains a formidable challenge, though key to understanding biological processes. We introduce Nellie, an automated and unbiased pipeline for segmentation, tracking, and feature extraction of diverse intracellular structures. Nellie adapts to image metadata, eliminating user input. Nellie's preprocessing pipeline enhances structural contrast on multiple intracellular scales allowing for robust hierarchical segmentation of sub-organellar regions. Internal motion capture markers are generated and tracked via a radius-adaptive pattern matching scheme, and used as guides for sub-voxel flow interpolation. Nellie extracts a plethora of features at multiple hierarchical levels for deep and customizable analysis. Nellie features a Napari-based GUI that allows for code-free operation and visualization, while its modular open-source codebase invites customization by experienced users. We demonstrate Nellie's wide variety of use cases with two examples: unmixing multiple organelles from a single channel using feature-based classification and training an unsupervised graph autoencoder on mitochondrial multi-mesh graphs to quantify latent space embedding changes following ionomycin treatment.
comment: for associated code, see https://github.com/aelefebv/nellie; 82 pages, 5 main figures, 11 extended figures
☆ Ax-to-Grind Urdu: Benchmark Dataset for Urdu Fake News Detection
Misinformation can seriously impact society, affecting anything from public opinion to institutional confidence and the political horizon of a state. Fake News (FN) proliferation on online websites and Online Social Networks (OSNs) has increased profusely. Various fact-checking websites include news in English and barely provide information about FN in regional languages. Thus the Urdu FN purveyors cannot be discerned using factchecking portals. SOTA approaches for Fake News Detection (FND) count upon appropriately labelled and large datasets. FND in regional and resource-constrained languages lags due to the lack of limited-sized datasets and legitimate lexical resources. The previous datasets for Urdu FND are limited-sized, domain-restricted, publicly unavailable and not manually verified where the news is translated from English into Urdu. In this paper, we curate and contribute the first largest publicly available dataset for Urdu FND, Ax-to-Grind Urdu, to bridge the identified gaps and limitations of existing Urdu datasets in the literature. It constitutes 10,083 fake and real news on fifteen domains collected from leading and authentic Urdu newspapers and news channel websites in Pakistan and India. FN for the Ax-to-Grind dataset is collected from websites and crowdsourcing. The dataset contains news items in Urdu from the year 2017 to the year 2023. Expert journalists annotated the dataset. We benchmark the dataset with an ensemble model of mBERT,XLNet, and XLM RoBERTa. The selected models are originally trained on multilingual large corpora. The results of the proposed model are based on performance metrics, F1-score, accuracy, precision, recall and MCC value.
☆ Searching Search Spaces: Meta-evolving a Geometric Encoding for Neural Networks
In evolutionary policy search, neural networks are usually represented using a direct mapping: each gene encodes one network weight. Indirect encoding methods, where each gene can encode for multiple weights, shorten the genome to reduce the dimensions of the search space and better exploit permutations and symmetries. The Geometric Encoding for Neural network Evolution (GENE) introduced an indirect encoding where the weight of a connection is computed as the (pseudo-)distance between the two linked neurons, leading to a genome size growing linearly with the number of genes instead of quadratically in direct encoding. However GENE still relies on hand-crafted distance functions with no prior optimization. Here we show that better performing distance functions can be found for GENE using Cartesian Genetic Programming (CGP) in a meta-evolution approach, hence optimizing the encoding to create a search space that is easier to exploit. We show that GENE with a learned function can outperform both direct encoding and the hand-crafted distances, generalizing on unseen problems, and we study how the encoding impacts neural network properties.
comment: 9 pages, 8 figures
☆ On Prompt Sensitivity of ChatGPT in Affective Computing
Recent studies have demonstrated the emerging capabilities of foundation models like ChatGPT in several fields, including affective computing. However, accessing these emerging capabilities is facilitated through prompt engineering. Despite the existence of some prompting techniques, the field is still rapidly evolving and many prompting ideas still require investigation. In this work, we introduce a method to evaluate and investigate the sensitivity of the performance of foundation models based on different prompts or generation parameters. We perform our evaluation on ChatGPT within the scope of affective computing on three major problems, namely sentiment analysis, toxicity detection, and sarcasm detection. First, we carry out a sensitivity analysis on pivotal parameters in auto-regressive text generation, specifically the temperature parameter $T$ and the top-$p$ parameter in Nucleus sampling, dictating how conservative or creative the model should be during generation. Furthermore, we explore the efficacy of several prompting ideas, where we explore how giving different incentives or structures affect the performance. Our evaluation takes into consideration performance measures on the affective computing tasks, and the effectiveness of the model to follow the stated instructions, hence generating easy-to-parse responses to be smoothly used in downstream applications.
comment: 2 Tables, 1 Figure, preprint submission to ACII 2024
☆ "This is not a data problem": Algorithms and Power in Public Higher Education in Canada
Algorithmic decision-making is increasingly being adopted across public higher education. The expansion of data-driven practices by post-secondary institutions has occurred in parallel with the adoption of New Public Management approaches by neoliberal administrations. In this study, we conduct a qualitative analysis of an in-depth ethnographic case study of data and algorithms in use at a public college in Ontario, Canada. We identify the data, algorithms, and outcomes in use at the college. We assess how the college's processes and relationships support those outcomes and the different stakeholders' perceptions of the college's data-driven systems. In addition, we find that the growing reliance on algorithmic decisions leads to increased student surveillance, exacerbation of existing inequities, and the automation of the faculty-student relationship. Finally, we identify a cycle of increased institutional power perpetuated by algorithmic decision-making, and driven by a push towards financial sustainability.
comment: In CHI '24 Proceedings of the CHI Conference on Human Factors in Computing Systems Honolulu, HI, USA
☆ Open Access NAO (OAN): a ROS2-based software framework for HRI applications with the NAO robot
This paper presents a new software framework for HRI experimentation with the sixth version of the common NAO robot produced by the United Robotics Group. Embracing the common demand of researchers for better performance and new features for NAO, the authors took advantage of the ability to run ROS2 onboard on the NAO to develop a framework independent of the APIs provided by the manufacturer. Such a system provides NAO with not only the basic skills of a humanoid robot such as walking and reproducing movements of interest but also features often used in HRI such as: speech recognition/synthesis, face and object detention, and the use of Generative Pre-trained Transformer (GPT) models for conversation. The developed code is therefore configured as a ready-to-use but also highly expandable and improvable tool thanks to the possibilities provided by the ROS community.
comment: 7 pages, 3 figures
☆ ACDG-VTON: Accurate and Contained Diffusion Generation for Virtual Try-On
Virtual Try-on (VTON) involves generating images of a person wearing selected garments. Diffusion-based methods, in particular, can create high-quality images, but they struggle to maintain the identities of the input garments. We identified this problem stems from the specifics in the training formulation for diffusion. To address this, we propose a unique training scheme that limits the scope in which diffusion is trained. We use a control image that perfectly aligns with the target image during training. In turn, this accurately preserves garment details during inference. We demonstrate our method not only effectively conserves garment details but also allows for layering, styling, and shoe try-on. Our method runs multi-garment try-on in a single inference cycle and can support high-quality zoomed-in generations without training in higher resolutions. Finally, we show our method surpasses prior methods in accuracy and quality.
☆ Evo* 2023 -- Late-Breaking Abstracts Volume
Volume with the Late-Breaking Abstracts submitted to the Evo* 2023 Conference, held in Brno (Czech Republic), from 12 to 14 of April. These papers present ongoing research and preliminary results investigating on the application of different approaches of Bioinspired Methods (mainly Evolutionary Computation) to different problems, most of them real world ones.
comment: LBAs accepted in Evo* 2023. Part of the Conference Proceedings
☆ BlendScape: Enabling Unified and Personalized Video-Conferencing Environments through Generative AI
Today's video-conferencing tools support a rich range of professional and social activities, but their generic, grid-based environments cannot be easily adapted to meet the varying needs of distributed collaborators. To enable end-user customization, we developed BlendScape, a system for meeting participants to compose video-conferencing environments tailored to their collaboration context by leveraging AI image generation techniques. BlendScape supports flexible representations of task spaces by blending users' physical or virtual backgrounds into unified environments and implements multimodal interaction techniques to steer the generation. Through an evaluation with 15 end-users, we investigated their customization preferences for work and social scenarios. Participants could rapidly express their design intentions with BlendScape and envisioned using the system to structure collaboration in future meetings, but experienced challenges with preventing distracting elements. We implement scenarios to demonstrate BlendScape's expressiveness in supporting distributed collaboration techniques from prior work and propose composition techniques to improve the quality of environments.
☆ Multi-criteria approach for selecting an explanation from the set of counterfactuals produced by an ensemble of explainers
Counterfactuals are widely used to explain ML model predictions by providing alternative scenarios for obtaining the more desired predictions. They can be generated by a variety of methods that optimize different, sometimes conflicting, quality measures and produce quite different solutions. However, choosing the most appropriate explanation method and one of the generated counterfactuals is not an easy task. Instead of forcing the user to test many different explanation methods and analysing conflicting solutions, in this paper, we propose to use a multi-stage ensemble approach that will select single counterfactual based on the multiple-criteria analysis. It offers a compromise solution that scores well on several popular quality measures. This approach exploits the dominance relation and the ideal point decision aid method, which selects one counterfactual from the Pareto front. The conducted experiments demonstrated that the proposed approach generates fully actionable counterfactuals with attractive compromise values of the considered quality measures.
comment: 17 pages, 2 figures
☆ Reducing Large Language Model Bias with Emphasis on 'Restricted Industries': Automated Dataset Augmentation and Prejudice Quantification
Despite the growing capabilities of large language models, there exists concerns about the biases they develop. In this paper, we propose a novel, automated mechanism for debiasing through specified dataset augmentation in the lens of bias producers and in the context of 'restricted industries' with limited data. We additionally create two new additional metrics, the mb-index and db-index, to quantify bias, considering the idea that bias occurs due to both intrinsic model architecture and dataset.
☆ Towards Learning Contrast Kinetics with Multi-Condition Latent Diffusion Models
Contrast agents in dynamic contrast enhanced magnetic resonance imaging allow to localize tumors and observe their contrast kinetics, which is essential for cancer characterization and respective treatment decision-making. However, contrast agent administration is not only associated with adverse health risks, but also restricted for patients during pregnancy, and for those with kidney malfunction, or other adverse reactions. With contrast uptake as key biomarker for lesion malignancy, cancer recurrence risk, and treatment response, it becomes pivotal to reduce the dependency on intravenous contrast agent administration. To this end, we propose a multi-conditional latent diffusion model capable of acquisition time-conditioned image synthesis of DCE-MRI temporal sequences. To evaluate medical image synthesis, we additionally propose and validate the Fr\'echet radiomics distance as an image quality measure based on biomarker variability between synthetic and real imaging data. Our results demonstrate our method's ability to generate realistic multi-sequence fat-saturated breast DCE-MRI and uncover the emerging potential of deep learning based contrast kinetics simulation. We publicly share our accessible codebase at https://github.com/RichardObi/ccnet.
☆ The Model Openness Framework: Promoting Completeness and Openness for Reproducibility, Transparency and Usability in AI
Generative AI (GAI) offers unprecedented possibilities but its commercialization has raised concerns about transparency, reproducibility, bias, and safety. Many "open-source" GAI models lack the necessary components for full understanding and reproduction, and some use restrictive licenses, a practice known as "openwashing." We propose the Model Openness Framework (MOF), a ranked classification system that rates machine learning models based on their completeness and openness, following principles of open science, open source, open data, and open access. The MOF requires specific components of the model development lifecycle to be included and released under appropriate open licenses. This framework aims to prevent misrepresentation of models claiming to be open, guide researchers and developers in providing all model components under permissive licenses, and help companies, academia, and hobbyists identify models that can be safely adopted without restrictions. Wide adoption of the MOF will foster a more open AI ecosystem, accelerating research, innovation, and adoption.
comment: 45 pages
♻ ☆ AnyHome: Open-Vocabulary Generation of Structured and Textured 3D Homes
Inspired by cognitive theories, we introduce AnyHome, a framework that translates any text into well-structured and textured indoor scenes at a house-scale. By prompting Large Language Models (LLMs) with designed templates, our approach converts provided textual narratives into amodal structured representations. These representations guarantee consistent and realistic spatial layouts by directing the synthesis of a geometry mesh within defined constraints. A Score Distillation Sampling process is then employed to refine the geometry, followed by an egocentric inpainting process that adds lifelike textures to it. AnyHome stands out with its editability, customizability, diversity, and realism. The structured representations for scenes allow for extensive editing at varying levels of granularity. Capable of interpreting texts ranging from simple labels to detailed narratives, AnyHome generates detailed geometries and textures that outperform existing methods in both quantitative and qualitative measures.
♻ ☆ Magic-Me: Identity-Specific Video Customized Diffusion
Creating content with specified identities (ID) has attracted significant interest in the field of generative models. In the field of text-to-image generation (T2I), subject-driven creation has achieved great progress with the identity controlled via reference images. However, its extension to video generation is not well explored. In this work, we propose a simple yet effective subject identity controllable video generation framework, termed Video Custom Diffusion (VCD). With a specified identity defined by a few images, VCD reinforces the identity characteristics and injects frame-wise correlation at the initialization stage for stable video outputs. To achieve this, we propose three novel components that are essential for high-quality identity preservation and stable video generation: 1) a noise initialization method with 3D Gaussian Noise Prior for better inter-frame stability; 2) an ID module based on extended Textual Inversion trained with the cropped identity to disentangle the ID information from the background 3) Face VCD and Tiled VCD modules to reinforce faces and upscale the video to higher resolution while preserving the identity's features. We conducted extensive experiments to verify that VCD is able to generate stable videos with better ID over the baselines. Besides, with the transferability of the encoded identity in the ID module, VCD is also working well with personalized text-to-image models available publicly. The codes are available at https://github.com/Zhen-Dong/Magic-Me.
comment: Project Page at https://magic-me-webpage.github.io
♻ ☆ Ada-NAV: Adaptive Trajectory Length-Based Sample Efficient Policy Learning for Robotic Navigation
Trajectory length stands as a crucial hyperparameter within reinforcement learning (RL) algorithms, significantly contributing to the sample inefficiency in robotics applications. Motivated by the pivotal role trajectory length plays in the training process, we introduce Ada-NAV, a novel adaptive trajectory length scheme designed to enhance the training sample efficiency of RL algorithms in robotic navigation tasks. Unlike traditional approaches that treat trajectory length as a fixed hyperparameter, we propose to dynamically adjust it based on the entropy of the underlying navigation policy. Interestingly, Ada-NAV can be applied to both existing on-policy and off-policy RL methods, which we demonstrate by empirically validating its efficacy on three popular RL methods: REINFORCE, Proximal Policy Optimization (PPO), and Soft Actor-Critic (SAC). We demonstrate through simulated and real-world robotic experiments that Ada-NAV outperforms conventional methods that employ constant or randomly sampled trajectory lengths. Specifically, for a fixed sample budget, Ada-NAV achieves an 18\% increase in navigation success rate, a 20-38\% reduction in navigation path length, and a 9.32\% decrease in elevation costs. Furthermore, we showcase the versatility of Ada-NAV by integrating it with the Clearpath Husky robot, illustrating its applicability in complex outdoor environments.
comment: 11 pages, 9 figures, 2 tables
♻ ☆ TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion Models
Despite remarkable achievements in video synthesis, achieving granular control over complex dynamics, such as nuanced movement among multiple interacting objects, still presents a significant hurdle for dynamic world modeling, compounded by the necessity to manage appearance and disappearance, drastic scale changes, and ensure consistency for instances across frames. These challenges hinder the development of video generation that can faithfully mimic real-world complexity, limiting utility for applications requiring high-level realism and controllability, including advanced scene simulation and training of perception systems. To address that, we propose TrackDiffusion, a novel video generation framework affording fine-grained trajectory-conditioned motion control via diffusion models, which facilitates the precise manipulation of the object trajectories and interactions, overcoming the prevalent limitation of scale and continuity disruptions. A pivotal component of TrackDiffusion is the instance enhancer, which explicitly ensures inter-frame consistency of multiple objects, a critical factor overlooked in the current literature. Moreover, we demonstrate that generated video sequences by our TrackDiffusion can be used as training data for visual perception models. To the best of our knowledge, this is the first work to apply video diffusion models with tracklet conditions and demonstrate that generated frames can be beneficial for improving the performance of object trackers.
♻ ☆ Having Beer after Prayer? Measuring Cultural Bias in Large Language Models
As the reach of large language models (LMs) expands globally, their ability to cater to diverse cultural contexts becomes crucial. Despite advancements in multilingual capabilities, models are not designed with appropriate cultural nuances. In this paper, we show that multilingual and Arabic monolingual LMs exhibit bias towards entities associated with Western culture. We introduce CAMeL, a novel resource of 628 naturally-occurring prompts and 20,368 entities spanning eight types that contrast Arab and Western cultures. CAMeL provides a foundation for measuring cultural biases in LMs through both extrinsic and intrinsic evaluations. Using CAMeL, we examine the cross-cultural performance in Arabic of 16 different LMs on tasks such as story generation, NER, and sentiment analysis, where we find concerning cases of stereotyping and cultural unfairness. We further test their text-infilling performance, revealing the incapability of appropriate adaptation to Arab cultural contexts. Finally, we analyze 6 Arabic pre-training corpora and find that commonly used sources such as Wikipedia may not be best suited to build culturally aware LMs, if used as they are without adjustment. We will make CAMeL publicly available at: https://github.com/tareknaous/camel
♻ ☆ Jaccard Metric Losses: Optimizing the Jaccard Index with Soft Labels NeurIPS 2023
Intersection over Union (IoU) losses are surrogates that directly optimize the Jaccard index. Leveraging IoU losses as part of the loss function have demonstrated superior performance in semantic segmentation tasks compared to optimizing pixel-wise losses such as the cross-entropy loss alone. However, we identify a lack of flexibility in these losses to support vital training techniques like label smoothing, knowledge distillation, and semi-supervised learning, mainly due to their inability to process soft labels. To address this, we introduce Jaccard Metric Losses (JMLs), which are identical to the soft Jaccard loss in standard settings with hard labels but are fully compatible with soft labels. We apply JMLs to three prominent use cases of soft labels: label smoothing, knowledge distillation and semi-supervised learning, and demonstrate their potential to enhance model accuracy and calibration. Our experiments show consistent improvements over the cross-entropy loss across 4 semantic segmentation datasets (Cityscapes, PASCAL VOC, ADE20K, DeepGlobe Land) and 13 architectures, including classic CNNs and recent vision transformers. Remarkably, our straightforward approach significantly outperforms state-of-the-art knowledge distillation and semi-supervised learning methods. The code is available at \href{https://github.com/zifuwanggg/JDTLosses}{https://github.com/zifuwanggg/JDTLosses}.
comment: NeurIPS 2023
♻ ☆ AutoMix: Automatically Mixing Language Models
Large language models (LLMs) are now available from cloud API providers in various sizes and configurations. While this diversity offers a broad spectrum of choices, effectively leveraging the options to optimize computational cost and performance remains challenging. In this work, we present AutoMix, an approach that strategically routes queries to larger LMs, based on the approximate correctness of outputs from a smaller LM. Central to AutoMix is a few-shot self-verification mechanism, which estimates the reliability of its own outputs without requiring training. Given that verifications can be noisy, we employ a meta-verifier in AutoMix to refine the accuracy of these assessments. Our experiments using LLAMA2-13B and GPT-4, on five context-grounded reasoning datasets demonstrate that AutoMix surpasses established baselines, improving the incremental benefit per cost by up to 86%. Our code and data are available at https://github.com/automix-llm/automix.
comment: The first two authors contributed equally. Work started and partly done during Aman's internship at Google. This version adds results on additional models and datasets
♻ ☆ Learning Algorithms for Verification of Markov Decision Processes
We present a general framework for applying learning algorithms and heuristical guidance to the verification of Markov decision processes (MDPs). The primary goal of our techniques is to improve performance by avoiding an exhaustive exploration of the state space, instead focussing on particularly relevant areas of the system, guided by heuristics. Our work builds on the previous results of Br{\'{a}}zdil et al., significantly extending it as well as refining several details and fixing errors. The presented framework focuses on probabilistic reachability, which is a core problem in verification, and is instantiated in two distinct scenarios. The first assumes that full knowledge of the MDP is available, in particular precise transition probabilities. It performs a heuristic-driven partial exploration of the model, yielding precise lower and upper bounds on the required probability. The second tackles the case where we may only sample the MDP without knowing the exact transition dynamics. Here, we obtain probabilistic guarantees, again in terms of both the lower and upper bounds, which provides efficient stopping criteria for the approximation. In particular, the latter is an extension of statistical model-checking (SMC) for unbounded properties in MDPs. In contrast to other related approaches, we do not restrict our attention to time-bounded (finite-horizon) or discounted properties, nor assume any particular structural properties of the MDP.
♻ ☆ MMICL: Empowering Vision-language Model with Multi-Modal In-Context Learning ICLR2024
Since the resurgence of deep learning, vision-language models (VLMs) enhanced by large language models (LLMs) have grown exponentially in popularity. However, while LLMs can utilize extensive background knowledge and task information with in-context learning, most VLMs still struggle with understanding complex multi-modal prompts with multiple images, making VLMs less effective in downstream vision-language tasks. In this paper, we address the limitation above by 1) introducing vision-language Model with Multi-Modal In-Context Learning(MMICL), a new approach to allow the VLM to deal with multi-modal inputs efficiently; 2) proposing a novel context scheme to augment the in-context learning ability of the VLM; 3) constructing the Multi-modal In-Context Learning (MIC) dataset, designed to enhance the VLM's ability to understand complex multi-modal prompts. Our experiments confirm that MMICL achieves new state-of-the-art zero-shot performance on a wide range of general vision-language tasks, especially for complex benchmarks, including MME and MMBench. Our analysis demonstrates that MMICL effectively tackles the challenge of complex multi-modal prompt understanding and emerges the impressive ICL ability. Furthermore, we observe that MMICL successfully alleviates language bias in VLMs, a common issue for VLMs that often leads to hallucination when faced with extensive textual context. Our code, dataset, dataset tool, and model are available at https://github.com/PKUnlp-icler/MIC
comment: Accepted by ICLR2024
♻ ☆ Graph Neural Networks for Learning Equivariant Representations of Neural Networks ICLR 2024
Neural networks that process the parameters of other neural networks find applications in domains as diverse as classifying implicit neural representations, generating neural network weights, and predicting generalization errors. However, existing approaches either overlook the inherent permutation symmetry in the neural network or rely on intricate weight-sharing patterns to achieve equivariance, while ignoring the impact of the network architecture itself. In this work, we propose to represent neural networks as computational graphs of parameters, which allows us to harness powerful graph neural networks and transformers that preserve permutation symmetry. Consequently, our approach enables a single model to encode neural computational graphs with diverse architectures. We showcase the effectiveness of our method on a wide range of tasks, including classification and editing of implicit neural representations, predicting generalization performance, and learning to optimize, while consistently outperforming state-of-the-art methods. The source code is open-sourced at https://github.com/mkofinas/neural-graphs.
comment: In ICLR 2024. Source code: https://github.com/mkofinas/neural-graphs
♻ ☆ Dice Semimetric Losses: Optimizing the Dice Score with Soft Labels MICCAI 2023
The soft Dice loss (SDL) has taken a pivotal role in numerous automated segmentation pipelines in the medical imaging community. Over the last years, some reasons behind its superior functioning have been uncovered and further optimizations have been explored. However, there is currently no implementation that supports its direct utilization in scenarios involving soft labels. Hence, a synergy between the use of SDL and research leveraging the use of soft labels, also in the context of model calibration, is still missing. In this work, we introduce Dice semimetric losses (DMLs), which (i) are by design identical to SDL in a standard setting with hard labels, but (ii) can be employed in settings with soft labels. Our experiments on the public QUBIQ, LiTS and KiTS benchmarks confirm the potential synergy of DMLs with soft labels (e.g. averaging, label smoothing, and knowledge distillation) over hard labels (e.g. majority voting and random selection). As a result, we obtain superior Dice scores and model calibration, which supports the wider adoption of DMLs in practice. The code is available at https://github.com/zifuwanggg/JDTLosses
comment: MICCAI 2023
♻ ☆ Observational and Experimental Insights into Machine Learning-Based Defect Classification in Wafers
This survey paper offers a comprehensive review of methodologies utilizing machine learning (ML) classification techniques for identifying wafer defects in semiconductor manufacturing. Despite the growing body of research demonstrating the effectiveness of ML in wafer defect identification, there is a noticeable absence of comprehensive reviews on this subject. This survey attempts to fill this void by amalgamating available literature and providing an in-depth analysis of the advantages, limitations, and potential applications of various ML classification algorithms in the realm of wafer defect detection. An innovative taxonomy of methodologies that we present provides a detailed classification of algorithms into more refined categories and techniques. This taxonomy follows a three-tier structure, starting from broad methodology categories and ending with specific techniques. It aids researchers in comprehending the complex relationships between different algorithms and their techniques. We employ a rigorous Observational and experimental evaluation to rank these varying techniques. For the Observational evaluation, we assess techniques based on a set of four criteria. The experimental evaluation ranks the algorithms employing the same techniques, sub-categories, and categories. Also the paper illuminates the future prospects of ML classification techniques for wafer defect identification, underscoring potential advancements and opportunities for further research in this field
♻ ☆ ABScribe: Rapid Exploration & Organization of Multiple Writing Variations in Human-AI Co-Writing Tasks using Large Language Models
Exploring alternative ideas by rewriting text is integral to the writing process. State-of-the-art Large Language Models (LLMs) can simplify writing variation generation. However, current interfaces pose challenges for simultaneous consideration of multiple variations: creating new variations without overwriting text can be difficult, and pasting them sequentially can clutter documents, increasing workload and disrupting writers' flow. To tackle this, we present ABScribe, an interface that supports rapid, yet visually structured, exploration and organization of writing variations in human-AI co-writing tasks. With ABScribe, users can swiftly modify variations using LLM prompts, which are auto-converted into reusable buttons. Variations are stored adjacently within text fields for rapid in-place comparisons using mouse-over interactions on a popup toolbar. Our user study with 12 writers shows that ABScribe significantly reduces task workload (d = 1.20, p < 0.001), enhances user perceptions of the revision process (d = 2.41, p < 0.001) compared to a popular baseline workflow, and provides insights into how writers explore variations using LLMs.
comment: CHI 2024
♻ ☆ Surfer: Progressive Reasoning with World Models for Robotic Manipulation
Considering how to make the model accurately understand and follow natural language instructions and perform actions consistent with world knowledge is a key challenge in robot manipulation. This mainly includes human fuzzy instruction reasoning and the following of physical knowledge. Therefore, the embodied intelligence agent must have the ability to model world knowledge from training data. However, most existing vision and language robot manipulation methods mainly operate in less realistic simulator and language settings and lack explicit modeling of world knowledge. To bridge this gap, we introduce a novel and simple robot manipulation framework, called Surfer. It is based on the world model, treats robot manipulation as a state transfer of the visual scene, and decouples it into two parts: action and scene. Then, the generalization ability of the model on new instructions and new scenes is enhanced by explicit modeling of the action and scene prediction in multi-modal information. In addition to the framework, we also built a robot manipulation simulator that supports full physics execution based on the MuJoCo physics engine. It can automatically generate demonstration training data and test data, effectively reducing labor costs. To conduct a comprehensive and systematic evaluation of the robot manipulation model in terms of language understanding and physical execution, we also created a robotic manipulation benchmark with progressive reasoning tasks, called SeaWave. It contains 4 levels of progressive reasoning tasks and can provide a standardized testing platform for embedded AI agents in multi-modal environments. On average, Surfer achieved a success rate of 54.74% on the defined four levels of manipulation tasks, exceeding the best baseline performance of 47.64%.
♻ ☆ LLM3:Large Language Model-based Task and Motion Planning with Motion Failure Reasoning IROS 2024
Conventional Task and Motion Planning (TAMP) approaches rely on manually crafted interfaces connecting symbolic task planning with continuous motion generation. These domain-specific and labor-intensive modules are limited in addressing emerging tasks in real-world settings. Here, we present LLM^3, a novel Large Language Model (LLM)-based TAMP framework featuring a domain-independent interface. Specifically, we leverage the powerful reasoning and planning capabilities of pre-trained LLMs to propose symbolic action sequences and select continuous action parameters for motion planning. Crucially, LLM^3 incorporates motion planning feedback through prompting, allowing the LLM to iteratively refine its proposals by reasoning about motion failure. Consequently, LLM^3 interfaces between task planning and motion planning, alleviating the intricate design process of handling domain-specific messages between them. Through a series of simulations in a box-packing domain, we quantitatively demonstrate the effectiveness of LLM^3 in solving TAMP problems and the efficiency in selecting action parameters. Ablation studies underscore the significant contribution of motion failure reasoning to the success of LLM^3. Furthermore, we conduct qualitative experiments on a physical manipulator, demonstrating the practical applicability of our approach in real-world settings.
comment: Submitted to IROS 2024. Codes available: https://github.com/AssassinWS/LLM-TAMP
♻ ☆ Do Language Models Know When They're Hallucinating References?
State-of-the-art language models (LMs) are notoriously susceptible to generating hallucinated information. Such inaccurate outputs not only undermine the reliability of these models but also limit their use and raise serious concerns about misinformation and propaganda. In this work, we focus on hallucinated book and article references and present them as the "model organism" of language model hallucination research, due to their frequent and easy-to-discern nature. We posit that if a language model cites a particular reference in its output, then it should ideally possess sufficient information about its authors and content, among other relevant details. Using this basic insight, we illustrate that one can identify hallucinated references without ever consulting any external resources, by asking a set of direct or indirect queries to the language model about the references. These queries can be considered as "consistency checks." Our findings highlight that while LMs, including GPT-4, often produce inconsistent author lists for hallucinated references, they also often accurately recall the authors of real references. In this sense, the LM can be said to "know" when it is hallucinating references. Furthermore, these findings show how hallucinated references can be dissected to shed light on their nature. Replication code and results can be found at https://github.com/microsoft/hallucinated-references.
♻ ☆ Vulnerability analysis of captcha using Deep learning
Several websites improve their security and avoid dangerous Internet attacks by implementing CAPTCHAs (Completely Automated Public Turing test to tell Computers and Humans Apart), a type of verification to identify whether the end-user is human or a robot. The most prevalent type of CAPTCHA is text-based, designed to be easily recognized by humans while being unsolvable towards machines or robots. However, as deep learning technology progresses, development of convolutional neural network (CNN) models that predict text-based CAPTCHAs becomes easier. The purpose of this research is to investigate the flaws and vulnerabilities in the CAPTCHA generating systems in order to design more resilient CAPTCHAs. To achieve this, we created CapNet, a Convolutional Neural Network. The proposed platform can evaluate both numerical and alphanumerical CAPTCHAs
♻ ☆ Analyzing and Improving the Training Dynamics of Diffusion Models
Diffusion models currently dominate the field of data-driven image synthesis with their unparalleled scaling to large datasets. In this paper, we identify and rectify several causes for uneven and ineffective training in the popular ADM diffusion model architecture, without altering its high-level structure. Observing uncontrolled magnitude changes and imbalances in both the network activations and weights over the course of training, we redesign the network layers to preserve activation, weight, and update magnitudes on expectation. We find that systematic application of this philosophy eliminates the observed drifts and imbalances, resulting in considerably better networks at equal computational complexity. Our modifications improve the previous record FID of 2.41 in ImageNet-512 synthesis to 1.81, achieved using fast deterministic sampling. As an independent contribution, we present a method for setting the exponential moving average (EMA) parameters post-hoc, i.e., after completing the training run. This allows precise tuning of EMA length without the cost of performing several training runs, and reveals its surprising interactions with network architecture, training time, and guidance.
♻ ☆ Real-Fake: Effective Training Data Synthesis Through Distribution Matching
Synthetic training data has gained prominence in numerous learning tasks and scenarios, offering advantages such as dataset augmentation, generalization evaluation, and privacy preservation. Despite these benefits, the efficiency of synthetic data generated by current methodologies remains inferior when training advanced deep models exclusively, limiting its practical utility. To address this challenge, we analyze the principles underlying training data synthesis for supervised learning and elucidate a principled theoretical framework from the distribution-matching perspective that explicates the mechanisms governing synthesis efficacy. Through extensive experiments, we demonstrate the effectiveness of our synthetic data across diverse image classification tasks, both as a replacement for and augmentation to real datasets, while also benefits such as out-of-distribution generalization, privacy preservation, and scalability. Specifically, we achieve 70.9% top1 classification accuracy on ImageNet1K when training solely with synthetic data equivalent to 1 X the original real data size, which increases to 76.0% when scaling up to 10 X synthetic data.
comment: Code released at (https://github.com/BAAI-DCAI/Training-Data-Synthesis)
♻ ☆ LogPrécis: Unleashing Language Models for Automated Shell Log Analysis
The collection of security-related logs holds the key to understanding attack behaviors and diagnosing vulnerabilities. Still, their analysis remains a daunting challenge. Recently, Language Models (LMs) have demonstrated unmatched potential in understanding natural and programming languages. The question arises whether and how LMs could be also useful for security experts since their logs contain intrinsically confused and obfuscated information. In this paper, we systematically study how to benefit from the state-of-the-art in LM to automatically analyze text-like Unix shell attack logs. We present a thorough design methodology that leads to LogPr\'ecis. It receives as input raw shell sessions and automatically identifies and assigns the attacker tactic to each portion of the session, i.e., unveiling the sequence of the attacker's goals. We demonstrate LogPr\'ecis capability to support the analysis of two large datasets containing about 400,000 unique Unix shell attacks. LogPr\'ecis reduces them into about 3,000 fingerprints, each grouping sessions with the same sequence of tactics. The abstraction it provides lets the analyst better understand attacks, identify fingerprints, detect novelty, link similar attacks, and track families and mutations. Overall, LogPr\'ecis, released as open source, paves the way for better and more responsive defense against cyberattacks.
comment: 18 pages, Computer&Security (https://www.sciencedirect.com/science/article/pii/S0167404824001068), code available at https://github.com/SmartData-Polito/logprecis, models available at https://huggingface.co/SmartDataPolito
♻ ☆ IVAC-P2L: Leveraging Irregular Repetition Priors for Improving Video Action Counting
Video Action Counting (VAC) is crucial in analyzing sports, fitness, and everyday activities by quantifying repetitive actions in videos. However, traditional VAC methods have overlooked the complexity of action repetitions, such as interruptions and the variability in cycle duration. Our research addresses the shortfall by introducing a novel approach to VAC, called Irregular Video Action Counting (IVAC). IVAC prioritizes modeling irregular repetition patterns in videos, which we define through two primary aspects: Inter-cycle Consistency and Cycle-interval Inconsistency. Inter-cycle Consistency ensures homogeneity in the spatial-temporal representations of cycle segments, signifying action uniformity within cycles. Cycle-interval inconsistency highlights the importance of distinguishing between cycle segments and intervals based on their inherent content differences. To encapsulate these principles, we propose a new methodology that includes consistency and inconsistency modules, supported by a unique pull-push loss (P2L) mechanism. The IVAC-P2L model applies a pull loss to promote coherence among cycle segment features and a push loss to clearly distinguish features of cycle segments from interval segments. Empirical evaluations conducted on the RepCount dataset demonstrate that the IVAC-P2L model sets a new benchmark in VAC task performance. Furthermore, the model demonstrates exceptional adaptability and generalization across various video contents, outperforming existing models on two additional datasets, UCFRep and Countix, without the need for dataset-specific optimization. These results confirm the efficacy of our approach in addressing irregular repetitions in videos and pave the way for further advancements in video analysis and understanding.
comment: Source code: https://github.com/hwang-cs-ime/IVAC-P2L
♻ ☆ Calibration of Deep Learning Classification Models in fNIRS
Functional near-infrared spectroscopy (fNIRS) is a valuable non-invasive tool for monitoring brain activity. The classification of fNIRS data in relation to conscious activity holds significance for advancing our understanding of the brain and facilitating the development of brain-computer interfaces (BCI). Many researchers have turned to deep learning to tackle the classification challenges inherent in fNIRS data due to its strong generalization and robustness. In the application of fNIRS, reliability is really important, and one mathematical formulation of the reliability of confidence is calibration. However, many researchers overlook the important issue of calibration. To address this gap, we propose integrating calibration into fNIRS field and assess the reliability of existing models. Surprisingly, our results indicate poor calibration performance in many proposed models. To advance calibration development in the fNIRS field, we summarize three practical tips. Through this letter, we hope to emphasize the critical role of calibration in fNIRS research and argue for enhancing the reliability of deep learning-based predictions in fNIRS classification tasks. All data from our experimental process are openly available on GitHub.
♻ ☆ MoST: Motion Style Transformer between Diverse Action Contents CVPR 2024
While existing motion style transfer methods are effective between two motions with identical content, their performance significantly diminishes when transferring style between motions with different contents. This challenge lies in the lack of clear separation between content and style of a motion. To tackle this challenge, we propose a novel motion style transformer that effectively disentangles style from content and generates a plausible motion with transferred style from a source motion. Our distinctive approach to achieving the goal of disentanglement is twofold: (1) a new architecture for motion style transformer with `part-attentive style modulator across body parts' and `Siamese encoders that encode style and content features separately'; (2) style disentanglement loss. Our method outperforms existing methods and demonstrates exceptionally high quality, particularly in motion pairs with different contents, without the need for heuristic post-processing. Codes are available at https://github.com/Boeun-Kim/MoST.
comment: Accepted by CVPR 2024
♻ ☆ The Power of Noise: Toward a Unified Multi-modal Knowledge Graph Representation Framework
The advancement of Multi-modal Pre-training highlights the necessity for a robust Multi-Modal Knowledge Graph (MMKG) representation learning framework. This framework is crucial for integrating structured knowledge into multi-modal Large Language Models (LLMs) at scale, aiming to alleviate issues like knowledge misconceptions and multi-modal hallucinations. In this work, to evaluate models' ability to accurately embed entities within MMKGs, we focus on two widely researched tasks: Multi-modal Knowledge Graph Completion (MKGC) and Multi-modal Entity Alignment (MMEA). Building on this foundation, we propose a novel SNAG method that utilizes a Transformer-based architecture equipped with modality-level noise masking for the robust integration of multi-modal entity features in KGs. By incorporating specific training objectives for both MKGC and MMEA, our approach achieves SOTA performance across a total of ten datasets (three for MKGC and seven for MEMA), demonstrating its robustness and versatility. Besides, SNAG can not only function as a standalone model but also enhance other existing methods, providing stable performance improvements. Our code and data are available at: https://github.com/zjukg/SNAG.
comment: Ongoing work; 10 pages, 6 Tables, 2 Figures; Repo is available at https://github.com/zjukg/SNAG
♻ ☆ BayesPrompt: Prompting Large-Scale Pre-Trained Language Models on Few-shot Inference via Debiased Domain Abstraction ICLR2024
As a novel and effective fine-tuning paradigm based on large-scale pre-trained language models (PLMs), prompt-tuning aims to reduce the gap between downstream tasks and pre-training objectives. While prompt-tuning has yielded continuous advancements in various tasks, such an approach still remains a persistent defect: prompt-tuning methods fail to generalize to specific few-shot patterns. From the perspective of distribution analyses, we disclose that the intrinsic issues behind the phenomenon are the over-multitudinous conceptual knowledge contained in PLMs and the abridged knowledge for target downstream domains, which jointly result in that PLMs mis-locate the knowledge distributions corresponding to the target domains in the universal knowledge embedding space. To this end, we intuitively explore to approximate the unabridged target domains of downstream tasks in a debiased manner, and then abstract such domains to generate discriminative prompts, thereby providing the de-ambiguous guidance for PLMs. Guided by such an intuition, we propose a simple yet effective approach, namely BayesPrompt, to learn prompts that contain the domain discriminative information against the interference from domain-irrelevant knowledge. BayesPrompt primitively leverages known distributions to approximate the debiased factual distributions of target domains and further uniformly samples certain representative features from the approximated distributions to generate the ultimate prompts for PLMs. We provide theoretical insights with the connection to domain adaptation. Empirically, our method achieves state-of-the-art performance on benchmarks.
comment: Accepted by ICLR2024
♻ ☆ MC-DBN: A Deep Belief Network-Based Model for Modality Completion
Recent advancements in multi-modal artificial intelligence (AI) have revolutionized the fields of stock market forecasting and heart rate monitoring. Utilizing diverse data sources can substantially improve prediction accuracy. Nonetheless, additional data may not always align with the original dataset. Interpolation methods are commonly utilized for handling missing values in modal data, though they may exhibit limitations in the context of sparse information. Addressing this challenge, we propose a Modality Completion Deep Belief Network-Based Model (MC-DBN). This approach utilizes implicit features of complete data to compensate for gaps between itself and additional incomplete data. It ensures that the enhanced multi-modal data closely aligns with the dynamic nature of the real world to enhance the effectiveness of the model. We conduct evaluations of the MC-DBN model in two datasets from the stock market forecasting and heart rate monitoring domains. Comprehensive experiments showcase the model's capacity to bridge the semantic divide present in multi-modal data, subsequently enhancing its performance. The source code is available at: https://github.com/logan-0623/DBN-generate
♻ ☆ CharPoet: A Chinese Classical Poetry Generation System Based on Token-free LLM
Automatic Chinese classical poetry generation has attracted much research interest, but achieving effective control over format and content simultaneously remains challenging. Traditional systems usually accept keywords as user inputs, resulting in limited control over content. Large language models (LLMs) improve content control by allowing unrestricted user instructions, but the token-by-token generation process frequently makes format errors. Motivated by this, we propose CharPoet, a Chinese classical poetry generation system based on token-free LLM, which provides effective control over both format and content. Our token-free architecture generates in a character-by-character manner, enabling precise control over the number of characters. Pruned from existing token-based LLMs, CharPoet inherits their pretrained capabilities and can generate poetry following instructions like "Write me a poem for my mother's birthday." CharPoet achieves format accuracy above 0.96, outperforming Jiuge-GPT-2 (0.91) and GPT-4 (0.38). In terms of content quality, CharPoet surpasses traditional systems including Jiuge, and is comparable to other LLMs. Our system is open source and available at https://modelscope.cn/models/CharPoet/CharPoet. A video demonstration of CharPoet is available at https://youtu.be/voZ25qEp3Dc.
♻ ☆ Mitigating Reversal Curse in Large Language Models via Semantic-aware Permutation Training
While large language models (LLMs) have achieved impressive performance across diverse tasks, recent studies showcase that causal LLMs suffer from the "reversal curse". It is a typical example that the model knows "A's father is B", but is unable to reason "B's child is A". This limitation poses a challenge to the advancement of artificial general intelligence (AGI), as it suggests a gap in the models' ability to comprehend and apply bidirectional reasoning. In this paper, we first conduct substantial evaluation and identify that the root cause of the reversal curse lies in the different word order between the training and inference stage, namely, the poor ability of causal language models to predict antecedent words within the training data. Accordingly, permutation on the training data is considered as a potential solution, since this can make the model predict antecedent words or tokens. However, previous permutation methods may disrupt complete phrases or entities, thereby posing challenges for the model to comprehend and learn from training data. To address this issue, we propose Semantic-aware Permutation Training (SPT), which addresses this issue by segmenting the training sentences into semantic units (i.e., entities or phrases) with an assistant language model and permuting these units before feeding into the model. Extensive experiments demonstrate that SPT effectively mitigates the reversal curse since the performance on reversed questions approximates that on the forward ones, and significantly advances the performance of existing works.
♻ ☆ A Pre-trained Data Deduplication Model based on Active Learning
In the era of big data, the issue of data quality has become increasingly prominent. One of the main challenges is the problem of duplicate data, which can arise from repeated entry or the merging of multiple data sources. These "dirty data" problems can significantly limit the effective application of big data. To address the issue of data deduplication, we propose a pre-trained deduplication model based on active learning, which is the first work that utilizes active learning to address the problem of deduplication at the semantic level. The model is built on a pre-trained Transformer and fine-tuned to solve the deduplication problem as a sequence to classification task, which firstly integrate the transformer with active learning into an end-to-end architecture to select the most valuable data for deduplication model training, and also firstly employ the R-Drop method to perform data augmentation on each round of labeled data, which can reduce the cost of manual labeling and improve the model's performance. Experimental results demonstrate that our proposed model outperforms previous state-of-the-art (SOTA) for deduplicated data identification, achieving up to a 28% improvement in Recall score on benchmark datasets.
♻ ☆ Learning from Label Proportions: Bootstrapping Supervised Learners via Belief Propagation ICLR 2024
Learning from Label Proportions (LLP) is a learning problem where only aggregate level labels are available for groups of instances, called bags, during training, and the aim is to get the best performance at the instance-level on the test data. This setting arises in domains like advertising and medicine due to privacy considerations. We propose a novel algorithmic framework for this problem that iteratively performs two main steps. For the first step (Pseudo Labeling) in every iteration, we define a Gibbs distribution over binary instance labels that incorporates a) covariate information through the constraint that instances with similar covariates should have similar labels and b) the bag level aggregated label. We then use Belief Propagation (BP) to marginalize the Gibbs distribution to obtain pseudo labels. In the second step (Embedding Refinement), we use the pseudo labels to provide supervision for a learner that yields a better embedding. Further, we iterate on the two steps again by using the second step's embeddings as new covariates for the next iteration. In the final iteration, a classifier is trained using the pseudo labels. Our algorithm displays strong gains against several SOTA baselines (up to 15%) for the LLP Binary Classification problem on various dataset types - tabular and Image. We achieve these improvements with minimal computational overhead above standard supervised learning due to Belief Propagation, for large bag sizes, even for a million samples.
comment: Published as a conference paper at The Twelfth International Conference on Learning Representations (ICLR 2024) & Oral Presentation at Regulatable ML @ NeurIPS 2023
♻ ☆ AdjointDPM: Adjoint Sensitivity Method for Gradient Backpropagation of Diffusion Probabilistic Models
Existing customization methods require access to multiple reference examples to align pre-trained diffusion probabilistic models (DPMs) with user-provided concepts. This paper aims to address the challenge of DPM customization when the only available supervision is a differentiable metric defined on the generated contents. Since the sampling procedure of DPMs involves recursive calls to the denoising UNet, na\"ive gradient backpropagation requires storing the intermediate states of all iterations, resulting in extremely high memory consumption. To overcome this issue, we propose a novel method AdjointDPM, which first generates new samples from diffusion models by solving the corresponding probability-flow ODEs. It then uses the adjoint sensitivity method to backpropagate the gradients of the loss to the models' parameters (including conditioning signals, network weights, and initial noises) by solving another augmented ODE. To reduce numerical errors in both the forward generation and gradient backpropagation processes, we further reparameterize the probability-flow ODE and augmented ODE as simple non-stiff ODEs using exponential integration. Finally, we demonstrate the effectiveness of AdjointDPM on three interesting tasks: converting visual effects into identification text embeddings, finetuning DPMs for specific types of stylization, and optimizing initial noise to generate adversarial samples for security auditing.
♻ ☆ Genixer: Empowering Multimodal Large Language Models as a Powerful Data Generator
Instruction tuning data is essential for training the Multimodal Large Language Models (MLLMs). However, the creation of high-quality instruction tuning data presents significant challenges. Prior methods that depended on GPT-4 for data generation were not only costly but also lacked satisfactory performance in complex tasks (i.e., grounding-based reasoning tasks). To address these issues, we developed an innovative data generation pipeline, Genixer, to generate various high-quality instruction tuning data, including nine representative tasks, e.g., Common VQA, REC, REG, and PointQ. Specifically, Genixer provides a unified solution with four key steps for alleviating the difficulty of data generation: (i) instruction data collection, (ii) instruction template design, (iii) empowering MLLM, and (iv) data generation and filtering. Subsequently, the superior qualitative results of our Genixer demonstrate that current MLLMs have a strong potential to evolve into powerful data generators. Additionally, to validate the efficacy of generated data quantitatively, we add the instruction tuning data produced by Genixer into the training of two representative MLLMs and observe the consistent improvements on various VQA tasks and multimodal benchmarks.
comment: Technical report
AgentOhana: Design Unified Data and Training Pipeline for Effective Agent Learning
Autonomous agents powered by large language models (LLMs) have garnered significant research attention. However, fully harnessing the potential of LLMs for agent-based tasks presents inherent challenges due to the heterogeneous nature of diverse data sources featuring multi-turn trajectories. In this paper, we introduce \textbf{AgentOhana} as a comprehensive solution to address these challenges. \textit{AgentOhana} aggregates agent trajectories from distinct environments, spanning a wide array of scenarios. It meticulously standardizes and unifies these trajectories into a consistent format, streamlining the creation of a generic data loader optimized for agent training. Leveraging the data unification, our training pipeline maintains equilibrium across different data sources and preserves independent randomness across devices during dataset partitioning and model training. Additionally, we present \textbf{xLAM-v0.1}, a large action model tailored for AI agents, which demonstrates exceptional performance across various benchmarks. Begin the exploration at \url{https://github.com/SalesforceAIResearch/xLAM}.
comment: Add GitHub repo link at \url{https://github.com/SalesforceAIResearch/xLAM} and HuggingFace model link at \url{https://huggingface.co/Salesforce/xLAM-v0.1-r}
♻ ☆ Vehicle Dispatching and Routing of On-Demand Intercity Ride-Pooling Services: A Multi-Agent Hierarchical Reinforcement Learning Approach
The integrated development of city clusters has given rise to an increasing demand for intercity travel. Intercity ride-pooling service exhibits considerable potential in upgrading traditional intercity bus services by implementing demand-responsive enhancements. Nevertheless, its online operations suffer the inherent complexities due to the coupling of vehicle resource allocation among cities and pooled-ride vehicle routing. To tackle these challenges, this study proposes a two-level framework designed to facilitate online fleet management. Specifically, a novel multi-agent feudal reinforcement learning model is proposed at the upper level of the framework to cooperatively assign idle vehicles to different intercity lines, while the lower level updates the routes of vehicles using an adaptive large neighborhood search heuristic. Numerical studies based on the realistic dataset of Xiamen and its surrounding cities in China show that the proposed framework effectively mitigates the supply and demand imbalances, and achieves significant improvement in both the average daily system profit and order fulfillment ratio.
♻ ☆ ERASE: Benchmarking Feature Selection Methods for Deep Recommender Systems
Deep Recommender Systems (DRS) are increasingly dependent on a large number of feature fields for more precise recommendations. Effective feature selection methods are consequently becoming critical for further enhancing the accuracy and optimizing storage efficiencies to align with the deployment demands. This research area, particularly in the context of DRS, is nascent and faces three core challenges. Firstly, variant experimental setups across research papers often yield unfair comparisons, obscuring practical insights. Secondly, the existing literature's lack of detailed analysis on selection attributes, based on large-scale datasets and a thorough comparison among selection techniques and DRS backbones, restricts the generalizability of findings and impedes deployment on DRS. Lastly, research often focuses on comparing the peak performance achievable by feature selection methods, an approach that is typically computationally infeasible for identifying the optimal hyperparameters and overlooks evaluating the robustness and stability of these methods. To bridge these gaps, this paper presents ERASE, a comprehensive bEnchmaRk for feAture SElection for DRS. ERASE comprises a thorough evaluation of eleven feature selection methods, covering both traditional and deep learning approaches, across four public datasets, private industrial datasets, and a real-world commercial platform, achieving significant enhancement. Our code is available online for ease of reproduction.
♻ ☆ LLatrieval: LLM-Verified Retrieval for Verifiable Generation NAACL 2024
Verifiable generation aims to let the large language model (LLM) generate text with supporting documents, which enables the user to flexibly verify the answer and makes the LLM's output more reliable. Retrieval plays a crucial role in verifiable generation. Specifically, the retrieved documents not only supplement knowledge to help the LLM generate correct answers, but also serve as supporting evidence for the user to verify the LLM's output. However, the widely used retrievers become the bottleneck of the entire pipeline and limit the overall performance. Their capabilities are usually inferior to LLMs since they often have much fewer parameters than the large language model and have not been demonstrated to scale well to the size of LLMs. If the retriever does not correctly find the supporting documents, the LLM can not generate the correct and verifiable answer, which overshadows the LLM's remarkable abilities. To address these limitations, we propose \LLatrieval (Large Language Model Verified Retrieval), where the LLM updates the retrieval result until it verifies that the retrieved documents can sufficiently support answering the question. Thus, the LLM can iteratively provide feedback to retrieval and facilitate the retrieval result to fully support verifiable generation. Experiments show that LLatrieval significantly outperforms extensive baselines and achieves state-of-the-art results.
comment: Accepted by NAACL 2024 (Main Conference)
♻ ☆ Exploring the Privacy-Energy Consumption Tradeoff for Split Federated Learning
Split Federated Learning (SFL) has recently emerged as a promising distributed learning technology, leveraging the strengths of both federated and split learning. It emphasizes the advantages of rapid convergence while addressing privacy concerns. As a result, this innovation has received significant attention from both industry and academia. However, since the model is split at a specific layer, known as a cut layer, into both client-side and server-side models for the SFL, the choice of the cut layer in SFL can have a substantial impact on the energy consumption of clients and their privacy, as it influences the training burden and the output of the client-side models. In this article, we provide a comprehensive overview of the SFL process and thoroughly analyze energy consumption and privacy. This analysis considers the influence of various system parameters on the cut layer selection strategy. Additionally, we provide an illustrative example of the cut layer selection, aiming to minimize clients' risk of reconstructing the raw data at the server while sustaining energy consumption within the required energy budget, which involves trade-offs. Finally, we address open challenges in this field. These directions represent promising avenues for future research and development.
comment: 10 pages, 5 figures
♻ ☆ Can It Edit? Evaluating the Ability of Large Language Models to Follow Code Editing Instructions
A significant amount of research is focused on developing and evaluating large language models for a variety of code synthesis tasks. These include synthesizing code from natural language, synthesizing tests from code, and synthesizing explanations of code. In contrast, the behavior of instructional code editing with LLMs is understudied. These are tasks in which the model is provided a block of code and an instruction to modify the code. The editing instruction may ask for a feature to be added or removed, describe a bug and ask for a fix, or ask for a different kind of solution. We introduce a carefully crafted benchmark of code editing tasks and use it to evaluate several cutting edge LLMs. Our evaluation exposes a significant gap between the capabilities of state-of-the-art open and closed models. For example, even GPT-3.5-Turbo is better than the best open model at code editing tasks. We also introduce a new, carefully curated, permissively licensed training dataset of code editing tasks coupled with natural language instructions. Using this training dataset, we show that we can fine-tune open Code LLMs to significantly improve their code editing capabilities, closing the gap between open and closed models. All code, data, and models are available at https://github.com/nuprl/CanItEdit.
♻ ☆ Exploring semantic information in disease: Simple Data Augmentation Techniques for Chinese Disease Normalization
Disease name normalization is an important task in the medical domain. It classifies disease names written in various formats into standardized names, serving as a fundamental component in smart healthcare systems for various disease-related functions. Nevertheless, the most significant obstacle to existing disease name normalization systems is the severe shortage of training data. While data augmentation is a powerful approach for addressing data scarcity, our findings reveal that conventional data augmentation techniques often impede task performance, primarily due to the multi-axis and multi-granularity nature of disease names. Consequently, we introduce a set of customized data augmentation techniques designed to leverage the semantic information inherent in disease names. These techniques aim to enhance the model's understanding of the semantic intricacies and classification structure of disease names. Through extensive experimentation, we illustrate that our proposed plug-and-play methods not only surpass general data augmentation techniques but also exhibit significant performance improvements across various baseline models and training objectives, particularly in scenarios with limited training data. This underscores its potential for widespread application in medical language processing tasks.
♻ ☆ Don't Blame the Annotator: Bias Already Starts in the Annotation Instructions EACL 2023
In recent years, progress in NLU has been driven by benchmarks. These benchmarks are typically collected by crowdsourcing, where annotators write examples based on annotation instructions crafted by dataset creators. In this work, we hypothesize that annotators pick up on patterns in the crowdsourcing instructions, which bias them to write many similar examples that are then over-represented in the collected data. We study this form of bias, termed instruction bias, in 14 recent NLU benchmarks, showing that instruction examples often exhibit concrete patterns, which are propagated by crowdworkers to the collected data. This extends previous work (Geva et al., 2019) and raises a new concern of whether we are modeling the dataset creator's instructions, rather than the task. Through a series of experiments, we show that, indeed, instruction bias can lead to overestimation of model performance, and that models struggle to generalize beyond biases originating in the crowdsourcing instructions. We further analyze the influence of instruction bias in terms of pattern frequency and model size, and derive concrete recommendations for creating future NLU benchmarks.
comment: EACL 2023 (Outstanding Paper Award)
♻ ☆ Content-aware Masked Image Modeling Transformer for Stereo Image Compression
Existing learning-based stereo image codec adopt sophisticated transformation with simple entropy models derived from single image codecs to encode latent representations. However, those entropy models struggle to effectively capture the spatial-disparity characteristics inherent in stereo images, which leads to suboptimal rate-distortion results. In this paper, we propose a stereo image compression framework, named CAMSIC. CAMSIC independently transforms each image to latent representation and employs a powerful decoder-free Transformer entropy model to capture both spatial and disparity dependencies, by introducing a novel content-aware masked image modeling (MIM) technique. Our content-aware MIM facilitates efficient bidirectional interaction between prior information and estimated tokens, which naturally obviates the need for an extra Transformer decoder. Experiments show that our stereo image codec achieves state-of-the-art rate-distortion performance on two stereo image datasets Cityscapes and InStereo2K with fast encoding and decoding speed.
♻ ☆ PAGE: Prototype-Based Model-Level Explanations for Graph Neural Networks AAAI-22
Aside from graph neural networks (GNNs) attracting significant attention as a powerful framework revolutionizing graph representation learning, there has been an increasing demand for explaining GNN models. Although various explanation methods for GNNs have been developed, most studies have focused on instance-level explanations, which produce explanations tailored to a given graph instance. In our study, we propose Prototype-bAsed GNN-Explainer (PAGE), a novel model-level GNN explanation method that explains what the underlying GNN model has learned for graph classification by discovering human-interpretable prototype graphs. Our method produces explanations for a given class, thus being capable of offering more concise and comprehensive explanations than those of instance-level explanations. First, PAGE selects embeddings of class-discriminative input graphs on the graph-level embedding space after clustering them. Then, PAGE discovers a common subgraph pattern by iteratively searching for high matching node tuples using node-level embeddings via a prototype scoring function, thereby yielding a prototype graph as our explanation. Using six graph classification datasets, we demonstrate that PAGE qualitatively and quantitatively outperforms the state-of-the-art model-level explanation method. We also carry out systematic experimental studies by demonstrating the relationship between PAGE and instance-level explanation methods, the robustness of PAGE to input data scarce environments, and the computational efficiency of the proposed prototype scoring function in PAGE.
comment: 18 pages, 12 figures, 5 tables; to appear in the IEEE Transactions on Pattern Analysis and Machine Intelligence (Please cite our journal version that will appear in an upcoming issue. Its two-page extended summary was presented in the AAAI-22 Student Abstract and Poster Program.)
♻ ☆ Calibrated Language Models Must Hallucinate STOC
Recent language models generate false but plausible-sounding text with surprising frequency. Such "hallucinations" are an obstacle to the usability of language-based AI systems and can harm people who rely upon their outputs. This work shows that there is an inherent statistical lower-bound on the rate that pretrained language models hallucinate certain types of facts, having nothing to do with the transformer LM architecture or data quality. For "arbitrary" facts whose veracity cannot be determined from the training data, we show that hallucinations must occur at a certain rate for language models that satisfy a statistical calibration condition appropriate for generative language models. Specifically, if the maximum probability of any fact is bounded, we show that the probability of generating a hallucination is close to the fraction of facts that occur exactly once in the training data (a "Good-Turing" estimate), even assuming ideal training data without errors. One conclusion is that models pretrained to be sufficiently good predictors (i.e., calibrated) may require post-training to mitigate hallucinations on the type of arbitrary facts that tend to appear once in the training set. However, our analysis also suggests that there is no statistical reason that pretraining will lead to hallucination on facts that tend to appear more than once in the training data (like references to publications such as articles and books, whose hallucinations have been particularly notable and problematic) or on systematic facts (like arithmetic calculations). Therefore, different architectures and learning algorithms may mitigate these latter types of hallucinations.
comment: In Proceedings of the 56th Annual ACM Symposium on Theory of Computing (STOC) 2024
♻ ☆ BAFFLE: Hiding Backdoors in Offline Reinforcement Learning Datasets
Reinforcement learning (RL) makes an agent learn from trial-and-error experiences gathered during the interaction with the environment. Recently, offline RL has become a popular RL paradigm because it saves the interactions with environments. In offline RL, data providers share large pre-collected datasets, and others can train high-quality agents without interacting with the environments. This paradigm has demonstrated effectiveness in critical tasks like robot control, autonomous driving, etc. However, less attention is paid to investigating the security threats to the offline RL system. This paper focuses on backdoor attacks, where some perturbations are added to the data (observations) such that given normal observations, the agent takes high-rewards actions, and low-reward actions on observations injected with triggers. In this paper, we propose Baffle (Backdoor Attack for Offline Reinforcement Learning), an approach that automatically implants backdoors to RL agents by poisoning the offline RL dataset, and evaluate how different offline RL algorithms react to this attack. Our experiments conducted on four tasks and four offline RL algorithms expose a disquieting fact: none of the existing offline RL algorithms is immune to such a backdoor attack. More specifically, Baffle modifies 10\% of the datasets for four tasks (3 robotic controls and 1 autonomous driving). Agents trained on the poisoned datasets perform well in normal settings. However, when triggers are presented, the agents' performance decreases drastically by 63.2\%, 53.9\%, 64.7\%, and 47.4\% in the four tasks on average. The backdoor still persists after fine-tuning poisoned agents on clean datasets. We further show that the inserted backdoor is also hard to be detected by a popular defensive method. This paper calls attention to developing more effective protection for the open-source offline RL dataset.
comment: Accepted at IEEE S&P (Oakland) 2024
♻ ☆ Advancing Beyond Identification: Multi-bit Watermark for Large Language Models NAACL 2024
We show the viability of tackling misuses of large language models beyond the identification of machine-generated text. While existing zero-bit watermark methods focus on detection only, some malicious misuses demand tracing the adversary user for counteracting them. To address this, we propose Multi-bit Watermark via Position Allocation, embedding traceable multi-bit information during language model generation. Through allocating tokens onto different parts of the messages, we embed longer messages in high corruption settings without added latency. By independently embedding sub-units of messages, the proposed method outperforms the existing works in terms of robustness and latency. Leveraging the benefits of zero-bit watermarking, our method enables robust extraction of the watermark without any model access, embedding and extraction of long messages ($\geq$ 32-bit) without finetuning, and maintaining text quality, while allowing zero-bit detection all at the same time. Code is released here: https://github.com/bangawayoo/mb-lm-watermarking
comment: NAACL 2024 main. 9 pages and appendix
♻ ☆ LingoQA: Video Question Answering for Autonomous Driving
Autonomous driving has long faced a challenge with public acceptance due to the lack of explainability in the decision-making process. Video question-answering (QA) in natural language provides the opportunity for bridging this gap. Nonetheless, evaluating the performance of Video QA models has proved particularly tough due to the absence of comprehensive benchmarks. To fill this gap, we introduce LingoQA, a benchmark specifically for autonomous driving Video QA. The LingoQA trainable metric demonstrates a 0.95 Spearman correlation coefficient with human evaluations. We introduce a Video QA dataset of central London consisting of 419k samples that we release with the paper. We establish a baseline vision-language model and run extensive ablation studies to understand its performance.
comment: Benchmark and dataset are available at https://github.com/wayveai/LingoQA/
♻ ☆ The Runtime of Random Local Search on the Generalized Needle Problem
In their recent work, C. Doerr and Krejca (Transactions on Evolutionary Computation, 2023) proved upper bounds on the expected runtime of the randomized local search heuristic on generalized Needle functions. Based on these upper bounds, they deduce in a not fully rigorous manner a drastic influence of the needle radius $k$ on the runtime. In this short article, we add the missing lower bound necessary to determine the influence of parameter $k$ on the runtime. To this aim, we derive an exact description of the expected runtime, which also significantly improves the upper bound given by C. Doerr and Krejca. We also describe asymptotic estimates of the expected runtime.
comment: 18 pages
♻ ☆ AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models ICLR 2024
The aligned Large Language Models (LLMs) are powerful language understanding and decision-making tools that are created through extensive alignment with human feedback. However, these large models remain susceptible to jailbreak attacks, where adversaries manipulate prompts to elicit malicious outputs that should not be given by aligned LLMs. Investigating jailbreak prompts can lead us to delve into the limitations of LLMs and further guide us to secure them. Unfortunately, existing jailbreak techniques suffer from either (1) scalability issues, where attacks heavily rely on manual crafting of prompts, or (2) stealthiness problems, as attacks depend on token-based algorithms to generate prompts that are often semantically meaningless, making them susceptible to detection through basic perplexity testing. In light of these challenges, we intend to answer this question: Can we develop an approach that can automatically generate stealthy jailbreak prompts? In this paper, we introduce AutoDAN, a novel jailbreak attack against aligned LLMs. AutoDAN can automatically generate stealthy jailbreak prompts by the carefully designed hierarchical genetic algorithm. Extensive evaluations demonstrate that AutoDAN not only automates the process while preserving semantic meaningfulness, but also demonstrates superior attack strength in cross-model transferability, and cross-sample universality compared with the baseline. Moreover, we also compare AutoDAN with perplexity-based defense methods and show that AutoDAN can bypass them effectively.
comment: Published as a conference paper at ICLR 2024. Code is available at https://github.com/SheltonLiu-N/AutoDAN
♻ ☆ Is this the real life? Is this just fantasy? The Misleading Success of Simulating Social Interactions With LLMs
Recent advances in large language models (LLM) have enabled richer social simulations, allowing for the study of various social phenomena with LLM-based agents. However, most work has used an omniscient perspective on these simulations (e.g., single LLM to generate all interlocutors), which is fundamentally at odds with the non-omniscient, information asymmetric interactions that humans have. To examine these differences, we develop an evaluation framework to simulate social interactions with LLMs in various settings (omniscient, non-omniscient). Our experiments show that interlocutors simulated omnisciently are much more successful at accomplishing social goals compared to non-omniscient agents, despite the latter being the more realistic setting. Furthermore, we demonstrate that learning from omniscient simulations improves the apparent naturalness of interactions but scarcely enhances goal achievement in cooperative scenarios. Our findings indicate that addressing information asymmetry remains a fundamental challenge for LLM-based agents.
♻ ☆ Metacognitive Prompting Improves Understanding in Large Language Models NAACL 2024
In Large Language Models (LLMs), there have been consistent advancements in task-specific performance, largely influenced by effective prompt design. Recent advancements in prompting have enhanced reasoning in logic-intensive tasks for LLMs, yet the nuanced understanding abilities of these models, crucial for processing and interpreting complex information, remain underexplored. In this study, we introduce Metacognitive Prompting (MP), a strategy inspired by human introspective reasoning processes. Using MP, LLMs undergo a systematic series of structured, self-aware evaluations, drawing on both their vast inherent knowledge and new insights. We conduct extensive experiments on four prevalent LLMs: Llama2, PaLM2, GPT-3.5, and GPT-4, across ten natural language understanding (NLU) datasets from GLUE, SuperGLUE, BLUE, and LexGLUE benchmarks. Additionally, we compare our method with chain-of-thought prompting and its advanced versions. The results show that GPT-4 consistently excels across all tasks, while other models have shown significant progress in some tasks when used in conjunction with MP. Furthermore, MP consistently outperforms existing prompting methods in both general and domain-specific NLU tasks. This study underscores the potential to amplify the understanding abilities of LLMs and highlights the benefits of mirroring human introspective reasoning in NLU tasks.
comment: NAACL 2024
♻ ☆ Beyond Specialization: Assessing the Capabilities of MLLMs in Age and Gender Estimation
Multimodal Large Language Models (MLLMs) have recently gained immense popularity. Powerful commercial models like ChatGPT-4V and Gemini, as well as open-source ones such as LLaVA, are essentially general-purpose models and are applied to solve a wide variety of tasks, including those in computer vision. These neural networks possess such strong general knowledge and reasoning abilities that they have proven capable of working even on tasks for which they were not specifically trained. We compared the capabilities of the most powerful MLLMs to date: ShareGPT4V, ChatGPT, LLaVA-Next in a specialized task of age and gender estimation with our state-of-the-art specialized model, MiVOLO. We also updated MiVOLO and provide details and new metrics in this article. This comparison has yielded some interesting results and insights about the strengths and weaknesses of the participating models. Furthermore, we attempted various ways to fine-tune the ShareGPT4V model for this specific task, aiming to achieve state-of-the-art results in this particular challenge. Although such a model would not be practical in production, as it is incredibly expensive compared to a specialized model like MiVOLO, it could be very useful in some tasks, like data annotation.
♻ ☆ Working Backwards: Learning to Place by Picking IROS'24
We present placing via picking (PvP), a method to autonomously collect real-world demonstrations for a family of placing tasks in which objects must be manipulated to specific contact-constrained locations. With PvP, we approach the collection of robotic object placement demonstrations by reversing the grasping process and exploiting the inherent symmetry of the pick and place problems. Specifically, we obtain placing demonstrations from a set of grasp sequences of objects initially located at their target placement locations. Our system can collect hundreds of demonstrations in contact-constrained environments without human intervention by combining two modules: tactile regrasping and compliant control for grasps. We train a policy directly from visual observations through behavioral cloning, using the autonomously-collected demonstrations. By doing so, the policy can generalize to object placement scenarios outside of the training environment without privileged information (e.g., placing a plate picked up from a table). We validate our approach in home robotic scenarios that include dishwasher loading and table setting. Our approach yields robotic placing policies that outperform policies trained with kinesthetic teaching, both in terms of performance and data efficiency, while requiring no human supervision.
comment: Submitted to the IEEE/RSJ International Conference on Intelligent Robotics and Systems (IROS'24), Abu Dhabi, UAE, Oct 14-18, 2024
♻ ☆ PhotoBot: Reference-Guided Interactive Photography via Natural Language IROS'24
We introduce PhotoBot, a framework for fully automated photo acquisition based on an interplay between high-level human language guidance and a robot photographer. We propose to communicate photography suggestions to the user via reference images that are selected from a curated gallery. We leverage a visual language model (VLM) and an object detector to characterize the reference images via textual descriptions and then use a large language model (LLM) to retrieve relevant reference images based on a user's language query through text-based reasoning. To correspond the reference image and the observed scene, we exploit pre-trained features from a vision transformer capable of capturing semantic similarity across marked appearance variations. Using these features, we compute pose adjustments for an RGB-D camera by solving a perspective-n-point (PnP) problem. We demonstrate our approach using a manipulator equipped with a wrist camera. Our user studies show that photos taken by PhotoBot are often more aesthetically pleasing than those taken by users themselves, as measured by human feedback. We also show that PhotoBot can generalize to other reference sources such as paintings.
comment: Submitted to the IEEE/RSJ International Conference on Intelligent Robotics and Systems (IROS'24), Abu Dhabi, UAE, Oct 14-18, 2024
♻ ☆ Discover and Mitigate Multiple Biased Subgroups in Image Classifiers CVPR 2024
Machine learning models can perform well on in-distribution data but often fail on biased subgroups that are underrepresented in the training data, hindering the robustness of models for reliable applications. Such subgroups are typically unknown due to the absence of subgroup labels. Discovering biased subgroups is the key to understanding models' failure modes and further improving models' robustness. Most previous works of subgroup discovery make an implicit assumption that models only underperform on a single biased subgroup, which does not hold on in-the-wild data where multiple biased subgroups exist. In this work, we propose Decomposition, Interpretation, and Mitigation (DIM), a novel method to address a more challenging but also more practical problem of discovering multiple biased subgroups in image classifiers. Our approach decomposes the image features into multiple components that represent multiple subgroups. This decomposition is achieved via a bilinear dimension reduction method, Partial Least Square (PLS), guided by useful supervision from the image classifier. We further interpret the semantic meaning of each subgroup component by generating natural language descriptions using vision-language foundation models. Finally, DIM mitigates multiple biased subgroups simultaneously via two strategies, including the data- and model-centric strategies. Extensive experiments on CIFAR-100 and Breeds datasets demonstrate the effectiveness of DIM in discovering and mitigating multiple biased subgroups. Furthermore, DIM uncovers the failure modes of the classifier on Hard ImageNet, showcasing its broader applicability to understanding model bias in image classifiers. The code is available at https://github.com/ZhangAIPI/DIM.
comment: CVPR 2024. Code is available at https://github.com/ZhangAIPI/DIM
♻ ☆ Graph Edits for Counterfactual Explanations: A comparative study
Counterfactuals have been established as a popular explainability technique which leverages a set of minimal edits to alter the prediction of a classifier. When considering conceptual counterfactuals on images, the edits requested should correspond to salient concepts present in the input data. At the same time, conceptual distances are defined by knowledge graphs, ensuring the optimality of conceptual edits. In this work, we extend previous endeavors on graph edits as counterfactual explanations by conducting a comparative study which encompasses both supervised and unsupervised Graph Neural Network (GNN) approaches. To this end, we pose the following significant research question: should we represent input data as graphs, which is the optimal GNN approach in terms of performance and time efficiency to generate minimal and meaningful counterfactual explanations for black-box image classifiers?
♻ ☆ Multi-Scale Contrastive Knowledge Co-Distillation for Event Temporal Relation Extraction
Event Temporal Relation Extraction (ETRE) is a crucial yet challenging problem. Event pairs are situated within a discourse at different distances, which we refer to as proximity bands. The temporal ordering communicated about event pairs situated at more remote (i.e., ``long'') or less remote (i.e., ``short'') proximity bands is encoded differently. SOTA ETRE models have tended to perform well on events situated at either short or long proximity bands, but not both. Yet, real-world, natural texts contain all types of temporal event-pairs. In this paper, we present MulCo: Multi-Scale Contrastive Knowledge Co-Distillation, a fusion approach that shares knowledge across multiple event pair proximity bands in order to improve performance on all types of temporal datasets. Our experimental results show that MulCo successfully integrates linguistic cues pertaining to temporal reasoning across both short and long proximity bands and achieves new state-of-the-art results on several ETRE benchmark datasets.
comment: update
♻ ☆ A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Ugly
Large Language Models (LLMs), such as ChatGPT and Bard, have revolutionized natural language understanding and generation. They possess deep language comprehension, human-like text generation capabilities, contextual awareness, and robust problem-solving skills, making them invaluable in various domains (e.g., search engines, customer support, translation). In the meantime, LLMs have also gained traction in the security community, revealing security vulnerabilities and showcasing their potential in security-related tasks. This paper explores the intersection of LLMs with security and privacy. Specifically, we investigate how LLMs positively impact security and privacy, potential risks and threats associated with their use, and inherent vulnerabilities within LLMs. Through a comprehensive literature review, the paper categorizes the papers into "The Good" (beneficial LLM applications), "The Bad" (offensive applications), and "The Ugly" (vulnerabilities of LLMs and their defenses). We have some interesting findings. For example, LLMs have proven to enhance code security (code vulnerability detection) and data privacy (data confidentiality protection), outperforming traditional methods. However, they can also be harnessed for various attacks (particularly user-level attacks) due to their human-like reasoning abilities. We have identified areas that require further research efforts. For example, Research on model and parameter extraction attacks is limited and often theoretical, hindered by LLM parameter scale and confidentiality. Safe instruction tuning, a recent development, requires more exploration. We hope that our work can shed light on the LLMs' potential to both bolster and jeopardize cybersecurity.
♻ ☆ Align before Adapt: Leveraging Entity-to-Region Alignments for Generalizable Video Action Recognition CVPR 2024
Large-scale visual-language pre-trained models have achieved significant success in various video tasks. However, most existing methods follow an "adapt then align" paradigm, which adapts pre-trained image encoders to model video-level representations and utilizes one-hot or text embedding of the action labels for supervision. This paradigm overlooks the challenge of mapping from static images to complicated activity concepts. In this paper, we propose a novel "Align before Adapt" (ALT) paradigm. Prior to adapting to video representation learning, we exploit the entity-to-region alignments for each frame. The alignments are fulfilled by matching the region-aware image embeddings to an offline-constructed text corpus. With the aligned entities, we feed their text embeddings to a transformer-based video adapter as the queries, which can help extract the semantics of the most important entities from a video to a vector. This paradigm reuses the visual-language alignment of VLP during adaptation and tries to explain an action by the underlying entities. This helps understand actions by bridging the gap with complex activity semantics, particularly when facing unfamiliar or unseen categories. ALT demonstrates competitive performance while maintaining remarkably low computational costs. In fully supervised experiments, it achieves 88.1% top-1 accuracy on Kinetics-400 with only 4947 GFLOPs. Moreover, ALT outperforms the previous state-of-the-art methods in both zero-shot and few-shot experiments, emphasizing its superior generalizability across various learning scenarios.
comment: Accepted at CVPR 2024
♻ ☆ Deep Reinforcement Learning with Hierarchical Reward Modeling
Reward design is a fundamental, yet challenging aspect of reinforcement learning (RL). Researchers typically utilize feedback signals from the environment to handcraft a reward function, but this process is not always effective due to the varying scale and intricate dependencies of the feedback signals. This paper shows by exploiting certain structures, one can ease the reward design process. Specifically, we propose a hierarchical reward modeling framework -- HERON for scenarios: (I) The feedback signals naturally present hierarchy; (II) The reward is sparse, but with less important surrogate feedback to help policy learning. Both scenarios allow us to design a hierarchical decision tree induced by the importance ranking of the feedback signals to compare RL trajectories. With such preference data, we can then train a reward model for policy learning. We apply HERON to several RL applications, and we find that our framework can not only train high performing agents on a variety of difficult tasks, but also provide additional benefits such as improved sample efficiency and robustness. Our code is available at \url{https://github.com/abukharin3/HERON}.
comment: 29 Pages, 15 figures
♻ ☆ RecMind: Large Language Model Powered Agent For Recommendation NAACL 2024
While the recommendation system (RS) has advanced significantly through deep learning, current RS approaches usually train and fine-tune models on task-specific datasets, limiting their generalizability to new recommendation tasks and their ability to leverage external knowledge due to model scale and data size constraints. Thus, we designed an LLM-powered autonomous recommender agent, RecMind, which is capable of leveraging external knowledge, utilizing tools with careful planning to provide zero-shot personalized recommendations. We propose a Self-Inspiring algorithm to improve the planning ability. At each intermediate step, the LLM self-inspires to consider all previously explored states to plan for the next step. This mechanism greatly improves the model's ability to comprehend and utilize historical information in planning for recommendation. We evaluate RecMind's performance in various recommendation scenarios. Our experiment shows that RecMind outperforms existing zero/few-shot LLM-based recommendation baseline methods in various tasks and achieves comparable performance to a fully trained recommendation model P5.
comment: Accepted by NAACL 2024 (Findings)
Optimization and Control 38
☆ Projection-free computation of robust controllable sets with constrained zonotopes
We study the problem of computing robust controllable sets for discrete-time linear systems with additive uncertainty. We propose a tractable and scalable approach to inner- and outer-approximate robust controllable sets using constrained zonotopes, when the additive uncertainty set is a symmetric, convex, and compact set. Our least-squares-based approach uses novel closed-form approximations of the Pontryagin difference between a constrained zonotopic minuend and a symmetric, convex, and compact subtrahend. Unlike existing approaches, our approach does not rely on convex optimization solvers, and is projection-free for ellipsoidal and zonotopic uncertainty sets. We also propose a least-squares-based approach to compute a convex, polyhedral outer-approximation to constrained zonotopes, and characterize sufficient conditions under which all these approximations are exact. We demonstrate the computational efficiency and scalability of our approach in several case studies, including the design of abort-safe rendezvous trajectories for a spacecraft in near-rectilinear halo orbit under uncertainty. Our approach can inner-approximate a 20-step robust controllable set for a 100-dimensional linear system in under 15 seconds on a standard computer.
comment: 22 pages, 6 figures
☆ Scalable Projection-Free Optimization Methods via MultiRadial Duality Theory
Recent works have developed new projection-free first-order methods based on utilizing linesearches and normal vector computations to maintain feasibility. These oracles can be cheaper than orthogonal projection or linear optimization subroutines but have the drawback of requiring a known strictly feasible point to do these linesearches with respect to. In this work, we develop new theory and algorithms which can operate using these cheaper linesearches while only requiring knowledge of points strictly satisfying each constraint separately. Convergence theory for several resulting ``multiradial'' gradient methods is established. We also provide preliminary numerics showing performance is essentially independent of how one selects the reference points for synthetic quadratically constrained quadratic programs.
comment: 26 pages
Optimal control of continuous-time symmetric systems with unknown dynamics and noisy measurements
An iterative learning algorithm is presented for continuous-time linear-quadratic optimal control problems where the system is externally symmetric with unknown dynamics. Both finite-horizon and infinite-horizon problems are considered. It is shown that the proposed algorithm is globally convergent to the optimal solution and has some advantages over adaptive dynamic programming, including being unbiased under noisy measurements and having a relatively low computational burden. Numerical experiments show the effectiveness of the results.
☆ Certified Constraint Propagation and Dual Proof Analysis in a Numerically Exact MIP Solver
This paper presents the integration of constraint propagation and dual proof analysis in an exact, roundoff-error-free MIP solver. The authors employ safe rounding methods to ensure that all results remain provably correct, while sacrificing as little computational performance as possible in comparison to a pure floating-point implementation. The study also addresses the adaptation of certification techniques for correctness verification. Computational studies demonstrate the effectiveness of these techniques, showcasing a 23% performance improvement on the MIPLIB 2017 benchmark test set.
☆ The defect, the Malgrange functor, and linear control systems
The notion of defect of a finitely presented functor on a module category is extended to arbitrary additive functors. The new defect and the contravariant Yoneda embedding form a right adjoint pair. The main result identifies the defect of the covariant Hom modulo projectives with the Bass torsion of the fixed argument. When applied to a linear control systems, it shows that the defect of the Malgrange functor of the system modulo projectives is isomorphic to the autonomy of the system. Furthermore, the defect of the contravariant Hom modulo injectives is shown to be isomorphic to the cotorsion coradical of the fixed argument. Since the Auslander-Gruson-Jensen transform of cotorsion is isomorphic to torsion, the above results raise two important questions: a) what is a control-theoretic interpretation of the covariant Yoneda embedding of the Malgrange module modulo injectives, and b) what is a control-theoretic interpretation of the Auslander-Gruson-Jensen duality?
comment: 11 pages. This text is an expanded version of Section 4 of an earlier version of the preprint entitled "The finite presentation of the stable Hom functors, the Bass torsion, and the cotorsion coradical". arXiv admin note: text overlap with arXiv:2212.09237
☆ Tikhonov regularized exterior penalty dynamics for constrained variational inequalities
Solving equilibrium problems under constraints is an important problem in optimization and optimal control. In this context an important practical challenge is the efficient incorporation of constraints. We develop a continuous-time method for solving constrained variational inequalities based on a new penalty regulated dynamical system in a general potentially infinite-dimensional Hilbert space. In order to obtain strong convergence of the issued trajectory of our method, we incorporate an explicit Tikhonov regularization parameter in our method, leading to a class of time-varying monotone inclusion problems featuring multiscale aspects. Besides strong convergence, we illustrate the practical efficiency of our developed method in solving constrained min-max problems.
comment: Submitted as a conference proceeding
☆ A multilevel framework for accelerating uSARA in radio-interferometric imaging
This paper presents a multilevel algorithm specifically designed for radio-interferometric imaging in astronomy. The proposed algorithm is used to solve the uSARA (unconstrained Sparsity Averaging Reweighting Analysis) formulation of this image restoration problem. Multilevel algorithms rely on a hierarchy of approximations of the objective function to accelerate its optimization. In contrast to the usual multilevel approaches where this hierarchy is derived in the parameter space, here we construct the hierarchy of approximations in the observation space. The proposed approach is compared to a reweighted forward-backward procedure, which is the backbone iteration scheme for solving the uSARA problem.
☆ Dynamic variable step size LMS adaptation algorithms -- Application to adaptive feedforward noise attenuation
The paper explores in detail the use of dynamic adaptation gain/step size (DAG) for improving the adaptation transient performance of variable step-size LMS (VS-LMS) adaptation algorithms. A generic form for the implementation of the DAG within the VS-LMS algorithms is provided. Criteria for the selection of the coefficients of the DAG filter which is required to be a strictly positive real transfer operator are given. The potential of the VS-LMS adaptation algorithms using a DAG is then illustrated by experimental results obtained on a relevant adaptive active noise attenuation system.
comment: European Control Conference 2024, Jun 2024, Stockholm, Sweden
☆ Direct search for stochastic optimization in random subspaces with zeroth-, first-, and second-order convergence and expected complexity
The work presented here is motivated by the development of StoDARS, a framework for large-scale stochastic blackbox optimization that not only is both an algorithmic and theoretical extension of the stochastic directional direct-search (SDDS) framework but also extends to noisy objectives a recent framework of direct-search algorithms in reduced spaces (DARS). Unlike SDDS, StoDARS achieves scalability by using~$m$ search directions generated in random subspaces defined through the columns of Johnson--Lindenstrauss transforms (JLTs) obtained from Haar-distributed orthogonal matrices. For theoretical needs, the quality of these subspaces and the accuracy of random estimates used by the algorithm are required to hold with sufficiently large, but fixed, probabilities. By leveraging an existing supermartingale-based framework, the expected complexity of StoDARS is proved to be similar to that of SDDS and other stochastic full-space methods up to constants, when the objective function is continuously differentiable. By dropping the latter assumption, the convergence of StoDARS to Clarke stationary points with probability one is established. Moreover, the analysis of the second-order behavior of the mesh adaptive direct-search (MADS) algorithm using a second-order-like extension of the Rademacher's theorem-based definition of the Clarke subdifferential (so-called generalized Hessian) is extended to the StoDARS framework, making it the first in a stochastic direct-search setting, to the best of our knowledge.
comment: 30 pages
☆ A Log-domain Interior Point Method for Convex Quadratic Games
In this paper, we propose an equilibrium-seeking algorithm for finding generalized Nash equilibria of non-cooperative monotone convex quadratic games. Specifically, we recast the Nash equilibrium-seeking problem as variational inequality problem that we solve using a log-domain interior point method and provide a general purpose solver based on this algorithm. This approach is suitable for non-potential, general sum games and does not require extensive structural assumptions. We demonstrate the efficiency and versatility of our method using three benchmark games and demonstrate our algorithm is especially effective on small to medium scale problems.
☆ Network-Aware Value Stacking of Community Battery via Asynchronous Distributed Optimization
Community battery systems have been widely deployed to provide services to the grid. Unlike a single battery storage system in the community, coordinating multiple community batteries can further unlock their value, enhancing the viability of community battery solutions. However, the centralized control of community batteries relies on the full information of the system, which is less practical and may even lead to privacy leakage. In this paper, we formulate a value-stacking optimization problem for community batteries to interact with local solar, buildings, and the grid, within distribution network constraints. We then propose a distributed algorithm using asynchronous distributed alternate direction method of multipliers (ADMM) to solve the problem. Our algorithm is robust to communication latency between community batteries and the grid while preserving the operational privacy. The simulation results demonstrate the convergence of our proposed asynchronous distributed ADMM algorithm. We also evaluate the electricity cost and the contribution of each value stream in the value-stacking problem for community batteries using real-world data.
comment: 2024 IEEE Power & Energy Society General Meeting (PESGM)
☆ Graph Attention Network-based Block Propagation with Optimal AoI and Reputation in Web 3.0
Web 3.0 is recognized as a pioneering paradigm that empowers users to securely oversee data without reliance on a centralized authority. Blockchains, as a core technology to realize Web 3.0, can facilitate decentralized and transparent data management. Nevertheless, the evolution of blockchain-enabled Web 3.0 is still in its nascent phase, grappling with challenges such as ensuring efficiency and reliability to enhance block propagation performance. In this paper, we design a Graph Attention Network (GAT)-based reliable block propagation optimization framework for blockchain-enabled Web 3.0. We first innovatively apply a data-freshness metric called age of information to measure block propagation efficiency in public blockchains. To achieve the reliability of block propagation, we introduce a reputation mechanism based on the subjective logic model, including the local and recommended opinions to calculate the miner reputation value. Moreover, considering that the GAT possesses the excellent ability to process graph-structured data, we utilize the GAT with reinforcement learning to obtain the optimal block propagation trajectory. Numerical results demonstrate that the proposed scheme exhibits the most outstanding block propagation efficiency and reliability compared with traditional routing algorithms.
☆ Safety-Aware Reinforcement Learning for Electric Vehicle Charging Station Management in Distribution Network
The increasing integration of electric vehicles (EVs) into the grid can pose a significant risk to the distribution system operation in the absence of coordination. In response to the need for effective coordination of EVs within the distribution network, this paper presents a safety-aware reinforcement learning (RL) algorithm designed to manage EV charging stations while ensuring the satisfaction of system constraints. Unlike existing methods, our proposed algorithm does not rely on explicit penalties for constraint violations, eliminating the need for penalty coefficient tuning. Furthermore, managing EV charging stations is further complicated by multiple uncertainties, notably the variability in solar energy generation and energy prices. To address this challenge, we develop an off-policy RL algorithm to efficiently utilize data to learn patterns in such uncertain environments. Our algorithm also incorporates a maximum entropy framework to enhance the RL algorithm's exploratory process, preventing convergence to local optimal solutions. Simulation results demonstrate that our algorithm outperforms traditional RL algorithms in managing EV charging in the distribution network.
comment: 2024 IEEE Power & Energy Society General Meeting (PESGM)
☆ Distributionally Robust Hospital Capacity Expansion Planning under Stochastic and Correlated Patient Demand
This paper investigates the optimal locations and capacities of hospital expansion facilities under uncertain future patient demands, considering both spatial and temporal correlations. We propose a novel two-stage distributionally robust optimization (DRO) model that integrates a Spatio-Temporal Neural Network (STNN). Specifically, we develop an STNN model that predicts future hospital occupancy levels considering spatial and temporal patterns in time-series datasets over a network of hospitals. The predictions of the STNN model are then used in the construction of the ambiguity set of the DRO model. To address computational challenges associated with two-stage DRO, we employ the linear-decision-rules technique to derive a tractable mixed-integer linear programming approximation. Extensive computational experiments conducted on real-world data demonstrate the superiority of the STNN model in minimizing forecast errors. Compared to neural network models built for each individual hospital, the proposed STNN model achieves a 53% improvement in average RMSE. Furthermore, the results demonstrate the value of incorporating spatiotemporal dependencies of demand uncertainty in the DRO model, as evidenced by out-of-sample analysis conducted with both ground truth data and under perfect information scenarios.
☆ Diffusion Model for Data-Driven Black-Box Optimization
Generative AI has redefined artificial intelligence, enabling the creation of innovative content and customized solutions that drive business practices into a new era of efficiency and creativity. In this paper, we focus on diffusion models, a powerful generative AI technology, and investigate their potential for black-box optimization over complex structured variables. Consider the practical scenario where one wants to optimize some structured design in a high-dimensional space, based on massive unlabeled data (representing design variables) and a small labeled dataset. We study two practical types of labels: 1) noisy measurements of a real-valued reward function and 2) human preference based on pairwise comparisons. The goal is to generate new designs that are near-optimal and preserve the designed latent structures. Our proposed method reformulates the design optimization problem into a conditional sampling problem, which allows us to leverage the power of diffusion models for modeling complex distributions. In particular, we propose a reward-directed conditional diffusion model, to be trained on the mixed data, for sampling a near-optimal solution conditioned on high predicted rewards. Theoretically, we establish sub-optimality error bounds for the generated designs. The sub-optimality gap nearly matches the optimal guarantee in off-policy bandits, demonstrating the efficiency of reward-directed diffusion models for black-box optimization. Moreover, when the data admits a low-dimensional latent subspace structure, our model efficiently generates high-fidelity designs that closely respect the latent structure. We provide empirical experiments validating our model in decision-making and content-creation tasks.
comment: arXiv admin note: substantial text overlap with arXiv:2307.07055
☆ Accelerated Objective Gap and Gradient Norm Convergence for Gradient Descent via Long Steps
This work considers gradient descent for L-smooth convex optimization with stepsizes larger than the classic regime where descent can be ensured. The stepsize schedules considered are similar to but differ slightly from the concurrently developed silver stepsizes of Altschuler and Parillo. For one of our stepsize sequences, we prove a $O\left(\frac{1}{N^{1.2716\dots}}\right)$ convergence rate in terms of objective gap decrease and for the other, we show the same rate of decrease for the squared-gradient-norm. This first result improves on the recent result of Altschuler and Parillo by a constant factor, while the second results improve on the exponent of the prior best squared-gradient-norm convergence guarantee of $O(\frac{1}{N})$.
☆ Performance-Guaranteed Solutions for Multi-Agent Optimal Coverage Problems using Submodularity, Curvature, and Greedy Algorithms
We consider a class of multi-agent optimal coverage problems where the goal is to determine the optimal placement of a group of agents in a given mission space such that they maximize a joint ``coverage'' objective. This class of problems is extremely challenging due to the non-convex nature of the mission space and of the coverage objective. With this motivation, we propose to use a greedy algorithm as a means of getting feasible coverage solutions efficiently. Even though such greedy solutions are suboptimal, the submodularity (diminishing returns) property of the coverage objective can be exploited to provide performance bound guarantees - not only for the greedy solutions but also for any subsequently improved solutions. Moreover, we show that improved performance bound guarantees (beyond the standard (1-1/e) performance bound) can be established using various curvature measures that further characterize the considered coverage problem. In particular, we provide a brief review of all existing popular curvature measures found in the submodular maximization literature, including a recent curvature measure that we proposed, and discuss in detail their applicability, practicality, and effectiveness in the context of optimal coverage problems. Moreover, we characterize the dependence of the effectiveness of different curvature measures (in providing improved performance bound guarantees) on the agent sensing capabilities. Finally, we provide several numerical results to support our findings and propose several potential future research directions.
comment: Will be submitted to CDC 2024
☆ Towards a connection between the capacitated vehicle routing problem and the constrained centroid-based clustering
Efficiently solving a vehicle routing problem (VRP) in a practical runtime is a critical challenge for delivery management companies. This paper explores both a theoretical and experimental connection between the Capacitated Vehicle Routing Problem (CVRP) and the Constrained Centroid-Based Clustering (CCBC). Reducing a CVRP to a CCBC is a synonym for a transition from an exponential to a polynomial complexity using commonly known algorithms for clustering, i.e K-means. At the beginning, we conduct an exploratory analysis to highlight the existence of such a relationship between the two problems through illustrative small-size examples and simultaneously deduce some mathematically-related formulations and properties. On a second level, the paper proposes a CCBC based approach endowed with some enhancements. The proposed framework consists of three stages. At the first step, a constrained centroid-based clustering algorithm generates feasible clusters of customers. This methodology incorporates three enhancement tools to achieve near-optimal clusters, namely: a multi-start procedure for initial centroids, a customer assignment metric, and a self-adjustment mechanism for choosing the number of clusters. At the second step, a traveling salesman problem (T SP) solver is used to optimize the order of customers within each cluster. Finally, we introduce a process relying on routes cutting and relinking procedure, which calls upon solving a linear and integer programming model to further improve the obtained routes. This step is inspired by the ruin & recreate algorithm. This approach is an extension of the classical cluster-first, route-second method and provides near-optimal solutions on well-known benchmark instances in terms of solution quality and computational runtime, offering a milestone in solving VRP.
☆ When are Lossy Energy Storage Optimization Models Convex?
We consider a class of optimization problems involving the optimal operation of a single lossy energy storage system that incurs energy loss when charging or discharging. Such inefficiencies in the energy storage dynamics are known to result in a nonconvex set of feasible charging and discharging power profiles. In this letter, we provide an equivalent reformulation for this class of optimization problems, along with sufficient conditions for the convexity of the proposed reformulation. The conditions provided generalize existing conditions for convexity in the literature.
comment: 5 pages, 1 figure
☆ High Accuracy Numerical Optimal Control for Rigid Bodies with Patch Contacts through Equivalent Contact Points -- Extended Version
This paper extends the Finite Elements with Switch Detection and Jumps (FESD-J) [1] method to problems of rigid body dynamics involving patch contacts. The FESD-J method is a high accuracy discretization scheme suitable for use in direct optimal control of nonsmooth mechanical systems. It detects dynamic switches exactly in time and, thereby, maintains the integration order of the underlying Runge- Kutta (RK) method. This is in contrast to commonly used time-stepping methods which only achieve first-order accuracy. Considering rigid bodies with possible patch contacts results in nondifferentiable signed distance functions (SDF), which introduces additional nonsmoothness into the dynamical system. In this work, we utilize so-called equivalent contact points (ECP), which parameterize force and impulse distributions on contact patches by evaluation at single points. We embed a nondifferentiable SDF into a complementarity Lagrangian system (CLS) and show that the determined ECP are well-defined. We then extend the FESD-J discretization to the considered CLS such that its integration accuracy is maintained. The functionality of the method is illustrated for both a simulation and an optimal control example.
comment: Shortened version submitted to 2024 Conference on Decision and Control (CDC)
Optimal Risk-Sensitive Scheduling Policies for Remote Estimation of Autoregressive Markov Processes
We design scheduling policies that minimize a risk-sensitive cost criterion for a remote estimation setup. Since risk-sensitive cost objective takes into account not just the mean value of the cost, but also higher order moments of its probability distribution, the resulting policy is robust to changes in the underlying system's parameters. The setup consists of a sensor that observes a discrete-time autoregressive Markov process, and at each time $t$ decides whether or not to transmit its observations to a remote estimator using an unreliable wireless communication channel after encoding these observations into data packets. We model the communication channel as a Gilbert-Elliott channel \cite{10384144}. Sensor probes the channel \cite{laourine2010betting} and hence knows the channel state at each time $t$ before making scheduling decision. The scheduler has to minimize the expected value of the exponential of the finite horizon cumulative cost that is sum of the following two quantities (i) the cumulative transmission power consumed, (ii) the cumulative squared estimator error. We pose this dynamic optimization problem as a Markov decision process (MDP), in which the system state at time $t$ is composed of (i) the instantaneous error $\Delta(t):= x(t)-a\hat{x}(t-1)$, where $x(t),\hat{x}(t-1)$ are the system state and the estimate at time $t,t-1$ respectively, and (ii) the channel state $c(t)$. We show that there exists an optimal policy that has a threshold structure, i.e., at each time $t$, for each possible channel state $c$, there is a threshold $\D\ust(c)$ such that if the current channel state is $c$, then it transmits only when the error $\D(t)$ exceeds $\D\ust(c)$.
☆ Analysing heavy-tail properties of Stochastic Gradient Descent by means of Stochastic Recurrence Equations
In recent works on the theory of machine learning, it has been observed that heavy tail properties of Stochastic Gradient Descent (SGD) can be studied in the probabilistic framework of stochastic recursions. In particular, G\"{u}rb\"{u}zbalaban et al. (arXiv:2006.04740) considered a setup corresponding to linear regression for which iterations of SGD can be modelled by a multivariate affine stochastic recursion $X_k=A_k X_{k-1}+B_k$, for independent and identically distributed pairs $(A_k, B_k)$, where $A_k$ is a random symmetric matrix and $B_k$ is a random vector. In this work, we will answer several open questions of the quoted paper and extend their results by applying the theory of irreducible-proximal (i-p) matrices.
comment: 25 pages, 2 figures
☆ A necessary condition for non-monotonic dose response, with an application to a kinetic proofreading model -- Extended version
Steady state non-monotonic ("biphasic") dose responses are often observed in experimental biology, which raises the control-theoretic question of identifying which possible mechanisms might underlie such behaviors. It is well known that the presence of an incoherent feedforward loop (IFFL) in a network may give rise to a non-monotonic response. It has been conjectured that this condition is also necessary, i.e. that a non-monotonic response implies the existence of an IFFL. In this paper, we show that this conjecture is false, and in the process prove a weaker version: that either an IFFL must exist or both a positive loop and a negative feedback loop must exist. Towards this aim, we give necessary and sufficient conditions for when minors of a symbolic matrix have mixed signs. Finally, we study in full generality when a model of immune T-cell activation could exhibit a steady state non-monotonic dose response.
comment: Appendix included
♻ ☆ Coupled linear Schrödinger equations: Control and stabilization results
This article presents some controllability and stabilization results for a system of two coupled linear Schr\"odinger equations in the one-dimensional case where the state components are interacting through the Kirchhoff boundary conditions. Considering the system in a bounded domain, the null boundary controllability result is shown. The result is achieved thanks to a new Carleman estimate, which ensures a boundary observation. Additionally, this boundary observation together with some trace estimates, helps us to use the Gramian approach, with a suitable choice of feedback law, to prove that the system under consideration decays exponentially to zero at least as fast as the function $e^{-2\omega t}$ for some $\omega>0$.
comment: 21 pp - Some new references and minor changes were made. Comments are welcome
♻ ☆ Stochastic Average Model Methods
We consider the solution of finite-sum minimization problems, such as those appearing in nonlinear least-squares or general empirical risk minimization problems. We are motivated by problems in which the summand functions are computationally expensive and evaluating all summands on every iteration of an optimization method may be undesirable. We present the idea of stochastic average model (SAM) methods, inspired by stochastic average gradient methods. SAM methods sample component functions on each iteration of a trust-region method according to a discrete probability distribution on component functions; the distribution is designed to minimize an upper bound on the variance of the resulting stochastic model. We present promising numerical results concerning an implemented variant extending the derivative-free model-based trust-region solver POUNDERS, which we name SAM-POUNDERS.
♻ ☆ Deep Reinforcement Learning: A Convex Optimization Approach
In this paper, we consider reinforcement learning of nonlinear systems with continuous state and action spaces. We present an episodic learning algorithm, where we for each episode use convex optimization to find a two-layer neural network approximation of the optimal $Q$-function. The convex optimization approach guarantees that the weights calculated at each episode are optimal, with respect to the given sampled states and actions of the current episode. For stable nonlinear systems, we show that the algorithm converges and that the converging parameters of the trained neural network can be made arbitrarily close to the optimal neural network parameters. In particular, if the regularization parameter is $\rho$ and the time horizon is $T$, then the parameters of the trained neural network converge to $w$, where the distance between $w$ from the optimal parameters $w^\star$ is bounded by $\mathcal{O}(\rho T^{-1})$. That is, when the number of episodes goes to infinity, there exists a constant $C$ such that \[\|w-w^\star\| \le C\cdot\frac{\rho}{T}.\] In particular, our algorithm converges arbitrarily close to the optimal neural network parameters as the time horizon increases or as the regularization parameter decreases.
♻ ☆ On Continuous Full-Order Integral-Terminal Sliding Mode Control with Unknown Apriori Bound on Uncertainty
This study aims at providing a solution to the problem of designing a continuous and finite-time control for a class of nonlinear systems in the presence of matched uncertainty with an unknown apriori bound. First, we propose a Full-Order Integral-Terminal Sliding Manifold (FOITSM) with a conventional (discontinuous) sliding mode to show that it provides the combined attributes of the nonsingular terminal and integral sliding mode algorithms. Secondly, an Adaptive Disturbance Observer (ADO) has been designed to alleviate the effect of the uncertainty acting on the system. On application of the ADO-based Full-Order Integral-Terminal Sliding Mode Control (FOITSMC), the chattering phenomenon in control input has been reduced substantially in the presence of conditionally known matched disturbances. Moreover, the adaptive gains of ADO are updated non-monotonically without over-bounding the acting disturbance, yet sustain the global boundedness of state trajectories within a specific bound. %Finally, an application of the proposed algorithm for attitude stabilization of a rigid spacecraft has been successively shown.
comment: 14 pages, 9 figures
♻ ☆ Optimal consumption and investment under relative performance criteria with Epstein-Zin utility
We consider the strategic interaction of traders in a continuous-time financial market with Epstein-Zin-type recursive intertemporal preferences and performance concerns. We derive explicitly an equilibrium for the finite player and the mean-field version of the game, based on a study of geometric backward stochastic differential equations of Bernoulli type that describe the best replies of traders. Our results show that Epstein-Zin preferences can lead to substantially different equilibrium behavior.
♻ ☆ PaPILO: A Parallel Presolving Library for Integer and Linear Programming with Multiprecision Support
Presolving has become an essential component of modern MIP solvers both in terms of computational performance and numerical robustness. In this paper, we present PaPILO, a new C++ header-only library that provides a large set of presolving routines for MIP and LP problems from the literature. The creation of PaPILO was motivated by the current lack of (a) solver-independent implementations that (b) exploit parallel hardware, and (c) support multiprecision arithmetic. Traditionally, presolving is designed to be fast. Whenever necessary, its low computational overhead is usually achieved by strict working limits. PaPILO's parallelization framework aims at reducing the computational overhead also when presolving is executed more aggressively or is applied to large-scale problems. To rule out conflicts between parallel presolve reductions, PaPILO uses a transaction-based design. This helps to avoid both the memory-intensive allocation of multiple copies of the problem and special synchronization between presolvers. Additionally, the use of Intel's TBB library aids PaPILO to efficiently exploit recursive parallelism within expensive presolving routines such as probing, dominated columns, or constraint sparsification. We provide an overview of PaPILO's capabilities and insights into important design choices.
♻ ☆ Certifying MIP-based Presolve Reductions for 0-1 Integer Linear Programs
It is well known that reformulating the original problem can be crucial for the performance of mixed-integer programming (MIP) solvers. To ensure correctness, all transformations must preserve the fea sibility status and optimal value of the problem, but there is currently no established methodology to express and verify the equivalence of two mixed-integer programs. In this work, we take a first step in this direction by showing how the correctness of MIP presolve reductions on 0-1 integer linear programs can be certified by using (and suitably extending) the VeriPB tool for pseudo-Boolean proof logging. Our experimental evaluation on both decision and optimization instances demonstrates the computational viability of the approach and leads to suggestions for future revisions of the proof format that will help to reduce the verbosity of the certificates and to accelerate the certification and verification process further.
♻ ☆ Optimization of inventory and capacity in large-scale assembly systems using extreme-value theory
High-tech systems are typically produced in two stages: 1) Production of components using specialized equipment and staff; 2) System assembly/integration. Component production capacity is subject to fluctuations, causing a high risk of shortages of at least one component, which results in costly delays. Companies hedge this risk by strategic investments in excess production capacity and in buffer inventories of components. To optimize these, it is crucial to characterize the relation between component shortage risk and capacity and inventory investments. We suppose that component production capacity and produce demand are normally distributed over finite time intervals, and we accordingly model the production system as a symmetric fork-join queueing network with $N$ statistically identical queues with a common arrival process and independent service processes. Assuming a symmetric cost structure, we subsequently apply extreme value theory to gain analytic insights into this optimization problem. We derive several new results for this queueing network, notably that the scaled maximum of $N$ steady-state queue lengths converges in distribution to a Gaussian random variable. These results translate into asymptotically optimal methods to dimension the system. Tests on a range of problems reveal that these methods typically work well for systems of moderate size.
♻ ☆ Supporting the next generation lithography roadmap using partial state-feedback reduced-order switching predictive models
To support the ever-increasing performance requirements of lithography systems in terms of throughput and accuracy, in this paper, we introduce a design framework for partial state-feedback reduced-order switching predictive models. By combining measurements and predictions, this method aims to: 1) improve overall system performance by reducing the placement errors in a die within and across the full-wafer and 2) eliminate redundant measurements by using predictions to improve system throughput. We primarily focus on well-known trade-off introduced by measurement time, which can correct errors at a cost of throughput, noise and not being robust to uncertainties. The proposed approach consists of a reduced-order model with a switching logic that acts a scheduler to deal with uncertain operating conditions. It also utilizes linear predictive models as a basis for the control design which is appealing to the ease and cost of implementation, enhancing the applicability. For the add-on part, the scheduler logic is adapted based on expected operating conditions of the system while guaranteeing global uniform ultimate bounded asymptotic stability. Lastly, to deal with measurement layouts, the predictor combines the measurement into model using partial state-feedback. Effectiveness of the proposed strategy is demonstrated in practice on a high-precision industrial scanner.
comment: 9 pages, 9 figures, to be submitted in IEEE-TIE
♻ ☆ A Simple Mixture Policy Parameterization for Improving Sample Efficiency of CVaR Optimization
Reinforcement learning algorithms utilizing policy gradients (PG) to optimize Conditional Value at Risk (CVaR) face significant challenges with sample inefficiency, hindering their practical applications. This inefficiency stems from two main facts: a focus on tail-end performance that overlooks many sampled trajectories, and the potential of gradient vanishing when the lower tail of the return distribution is overly flat. To address these challenges, we propose a simple mixture policy parameterization. This method integrates a risk-neutral policy with an adjustable policy to form a risk-averse policy. By employing this strategy, all collected trajectories can be utilized for policy updating, and the issue of vanishing gradients is counteracted by stimulating higher returns through the risk-neutral component, thus lifting the tail and preventing flatness. Our empirical study reveals that this mixture parameterization is uniquely effective across a variety of benchmark domains. Specifically, it excels in identifying risk-averse CVaR policies in some Mujoco environments where the traditional CVaR-PG fails to learn a reasonable policy.
♻ ☆ Tropical modeling of battery swapping and charging station
We propose and investigate a queueing model of a battery swapping and charging station (BSCS) for electric vehicles (EVs). A new approach to the analysis of the queueing model is developed, which combines the representation of the model as a stochastic dynamic system with the use of the methods and results of tropical algebra, which deals with the theory and applications of algebraic systems with idempotent operations. We describe the dynamics of the queueing model by a system of recurrence equations that involve random variables (RVs) to represent the interarrival time of incoming EVs. A performance measure for the model is defined as the mean operation cycle time of the station. Furthermore, the system of equations is represented in terms of the tropical algebra in vector form as an implicit linear state dynamic equation. The performance measure takes on the meaning of the mean growth rate of the state vector (the Lyapunov exponent) of the dynamic system. By applying a solution technique of vector equations in tropical algebra, the implicit equation is transformed into an explicit one with a state transition matrix with random entries. The evaluation of the Lyapunov exponent reduces to finding the limit of the expected value of norms of tropical matrix products. This limit is then obtained using results from the tropical spectral theory of deterministic and random matrices. With this approach, we derive a new exact formula for the mean cycle time of the BSCS, which is given in terms of the expected value of the RVs involved. We present the results of the Monte Carlo simulation of the BSCS's operation, which show a good agreement with the exact solution. The application of the obtained solution to evaluate the performance of one BSCS and to find the optimal distribution of battery packs between stations in a network of BSCSs is discussed.
comment: 21 pages, 5 figures
♻ ☆ On semidefinite descriptions for convex hulls of quadratic programs
Quadratically constrained quadratic programs (QCQPs) are a highly expressive class of nonconvex optimization problems. While QCQPs are NP-hard in general, they admit a natural convex relaxation via the standard semidefinite program (SDP) relaxation. In this paper we study when the convex hull of the epigraph of a QCQP coincides with the projected epigraph of the SDP relaxation. We present a sufficient condition for convex hull exactness and show that this condition is further necessary under an additional geometric assumption. The sufficient condition is based on geometric properties of $\Gamma$, the cone of convex Lagrange multipliers, and its relatives $\Gamma_1$ and $\Gamma^\circ$.
comment: This paper is a significant rewrite of arXiv:2011.07155 [math.OC] and contains both new content and rewritten content
♻ ☆ Application of tropical optimization for solving multicriteria problems of pairwise comparisons using log-Chebyshev approximation
We consider a decision-making problem to find absolute ratings of alternatives that are compared in pairs under multiple criteria, subject to constraints in the form of two-sided bounds on ratios between the ratings. Given matrices of pairwise comparisons made according to the criteria, the problem is formulated as the log-Chebyshev approximation of these matrices by a common consistent matrix (a symmetrically reciprocal matrix of unit rank) to minimize the approximation errors for all matrices simultaneously. We rearrange the approximation problem as a constrained multiobjective optimization problem of finding a vector that determines the approximating consistent matrix. The problem is then represented in the framework of tropical algebra, which deals with the theory and applications of idempotent semirings and provides a formal basis for fuzzy and interval arithmetic. We apply methods and results of tropical optimization to develop a new approach for handling the multiobjective optimization problem according to various principles of optimality. New complete solutions in the sense of the max-ordering, lexicographic ordering and lexicographic max-ordering optimality are obtained, which are given in a compact vector form ready for formal analysis and efficient computation. We present numerical examples of solving multicriteria problems of rating four alternatives from pairwise comparisons to illustrate the technique and compare it with others.
comment: 44 pages
♻ ☆ Ahlfors regularity of continua that minimize maxitive set functions
The primary objective of this paper is to establish the Ahlfors regularity of minimizers of set functions that satisfy a suitable maxitive condition on disjoint unions of sets. Our analysis focuses on minimizers within continua of the plane with finite one-dimensional Hausdorff measure. Through quantitative estimates, we prove that the length of a minimizer inside the ball centered at one of its points is comparable to the radius of the ball. By operating within an abstract framework, we are able to encompass a diverse range of entities, including spectral functionals defined in terms of the eigenvalues of elliptic operators, the inradius, and the maximum of the torsion function. These entities are of interest for several applications, such as structural engineering, urban planning, and quantum mechanics.
♻ ☆ Parallelized Conflict Graph Cut Generation
A conflict graph represents logical relations between binary variables, and effective use of the graph can significantly accelerate branch-and-cut solvers for mixed-integer programming (MIP). In this paper we develop efficient parallel conflict graph management: conflict detection; maximal clique generation; clique extension; and clique merging. We leverage parallel computing in order to intensify computational effort on the conflict graph, thereby generating a much larger pool of cutting planes than what can be practically achieved in serial. Computational experiments demonstrate that the expanded pool of cuts enabled by parallel computing lead to substantial reductions in total MIP solve time, especially for more challenging cases.
comment: 19 pages, 2 figures
Systems and Control 42
☆ Projection-free computation of robust controllable sets with constrained zonotopes
We study the problem of computing robust controllable sets for discrete-time linear systems with additive uncertainty. We propose a tractable and scalable approach to inner- and outer-approximate robust controllable sets using constrained zonotopes, when the additive uncertainty set is a symmetric, convex, and compact set. Our least-squares-based approach uses novel closed-form approximations of the Pontryagin difference between a constrained zonotopic minuend and a symmetric, convex, and compact subtrahend. Unlike existing approaches, our approach does not rely on convex optimization solvers, and is projection-free for ellipsoidal and zonotopic uncertainty sets. We also propose a least-squares-based approach to compute a convex, polyhedral outer-approximation to constrained zonotopes, and characterize sufficient conditions under which all these approximations are exact. We demonstrate the computational efficiency and scalability of our approach in several case studies, including the design of abort-safe rendezvous trajectories for a spacecraft in near-rectilinear halo orbit under uncertainty. Our approach can inner-approximate a 20-step robust controllable set for a 100-dimensional linear system in under 15 seconds on a standard computer.
comment: 22 pages, 6 figures
☆ On Optimal Management of Energy Storage Systems in Renewable Energy Communities
Renewable energy communities are legal entities involving the association of citizens, organizations and local businesses aimed at contributing to the green energy transition and providing social, environmental and economic benefits to their members. This goal is pursued through the cooperative efforts of the community actors and by increasing the local energy self-consumption. In this paper, the optimal energy community operation in the presence of energy storage units is addressed. By exploiting the flexibility provided by the storage facilities, the main task is to minimize the community energy bill by taking advantage of incentives related to local self-consumption. Optimality conditions are derived, and an explicit optimal solution is devised. Numerical simulations are provided to assess the performance of the proposed solution.
☆ Adaptive Reconstruction of Nonlinear Systems States via DREM with Perturbation Annihilation
A new adaptive observer is proposed for a certain class of nonlinear systems with bounded unknown input and parametric uncertainty. Unlike most existing solutions, the proposed approach ensures asymptotic convergence of the unknown parameters, state and perturbation estimates to an arbitrarily small neighborhood of the equilibrium point. The solution is based on the novel augmentation of a high-gain observer with the dynamic regressor extension and mixing (DREM) procedure enhanced with a perturbation annihilation algorithm. The aforementioned properties of the proposed solution are verified via numerical experiments.
comment: 6 pages, 2 figures
☆ Macroscopic pricing schemes for the utilization of pool ride-hailing vehicles in bus lanes
With the increasing popularity of ride-hailing services, new modes of transportation are having a significant impact on the overall performance of transportation networks. As a result, there is a need to ensure that both the various transportation alternatives and the spatial network resources are used efficiently. In this work, we analyze a network configuration where part of the urban transportation network is devoted to dedicated bus lanes. Apart from buses, we let pool ride-hailing trips use the dedicated bus lanes which, contingent upon the demand for the remaining modes, may result in faster trips for users opting for the pooling alternative. Under an aggregated modelling framework, we characterize the spatial configuration and the multi-modal demand split for which this strategy achieves a system optimum. For these specific scenarios, we compute the equilibrium when ride-hailing users can choose between solo and pool services, and we provide a pricing scheme for mitigating the gap between total user delays of the system optimum and user equilibrium solutions, when needed.
comment: Extended version of paper accepted to European Control Conference (ECC) 2024
☆ Priority-based Energy Allocation in Buildings for Distributed Model Predictive Control
Many countries are facing energy shortage today and most of the global energy is consumed by HVAC systems in buildings. For the scenarios where the energy system is not sufficiently supplied to HVAC systems, a priority-based allocation scheme based on distributed model predictive control is proposed in this paper, which distributes the energy rationally based on priority order. According to the scenarios, two distributed allocation strategies, i.e., one-to-one priority strategy and multi-to-one priority strategy, are developed in this paper and validated by simulation in a building containing three zones and a building containing 36 rooms, respectively. Both strategies fully exploit the potential of predictive control solutions. The experiment shows that our scheme has good scalability and achieve the performance of centralized strategy while making the calculation tractable.
☆ 3D Directed Formation Control with Global Shape Convergence using Bispherical Coordinates
In this paper, we present a novel 3D formation control scheme for directed graphs in a leader-follower configuration, achieving (almost) global convergence to the desired shape. Specifically, we introduce three controlled variables representing bispherical coordinates that uniquely describe the formation in 3D. Acyclic triangulated directed graphs (a class of minimally acyclic persistent graphs) are used to model the inter-agent sensing topology, while the agents' dynamics are governed by single-integrator model. Our analysis demonstrates that the proposed decentralized formation controller ensures (almost) global asymptotic stability while avoiding potential shape ambiguities in the final formation. Furthermore, the control laws are implementable in arbitrarily oriented local coordinate frames of follower agents using only low-cost onboard vision sensors, making it suitable for practical applications. Finally, we validate our formation control approach by a simulation study.
comment: Submitted to ECC 2024
Optimal control of continuous-time symmetric systems with unknown dynamics and noisy measurements
An iterative learning algorithm is presented for continuous-time linear-quadratic optimal control problems where the system is externally symmetric with unknown dynamics. Both finite-horizon and infinite-horizon problems are considered. It is shown that the proposed algorithm is globally convergent to the optimal solution and has some advantages over adaptive dynamic programming, including being unbiased under noisy measurements and having a relatively low computational burden. Numerical experiments show the effectiveness of the results.
☆ Bayesian Physics-informed Neural Networks for System Identification of Inverter-dominated Power Systems
While the uncertainty in generation and demand increases, accurately estimating the dynamic characteristics of power systems becomes crucial for employing the appropriate control actions to maintain their stability. In our previous work, we have shown that Bayesian Physics-informed Neural Networks (BPINNs) outperform conventional system identification methods in identifying the power system dynamic behavior under measurement noise. This paper takes the next natural step and addresses the more significant challenge, exploring how BPINN perform in estimating power system dynamics under increasing uncertainty from many Inverter-based Resources (IBRs) connected to the grid. These introduce a different type of uncertainty, compared to noisy measurements. The BPINN combines the advantages of Physics-informed Neural Networks (PINNs), such as inverse problem applicability, with Bayesian approaches for uncertainty quantification. We explore the BPINN performance on a wide range of systems, starting from a single machine infinite bus (SMIB) system and 3-bus system to extract important insights, to the 14-bus CIGRE distribution grid, and the large IEEE 118-bus system. We also investigate approaches that can accelerate the BPINN training, such as pretraining and transfer learning. Throughout this paper, we show that in presence of uncertainty, the BPINN achieves orders of magnitude lower errors than the widely popular method for system identification SINDy and significantly lower errors than PINN, while transfer learning helps reduce training time by up to 80 %.
comment: Submitted to Electric Power Systems Research
☆ Lattice piecewise affine approximation of explicit model predictive control with application to satellite attitude control
Satellite attitude cotrol is a crucial part of aerospace technology, and model predictive control(MPC) is one of the most promising controllers in this area, which will be less effective if real-time online optimization can not be achieved. Explicit MPC converts the online calculation into a table lookup process, however the solution is difficult to obtain if the system dimension is high or the constraints are complex. The lattice piecewise affine(PWA) function was used to represent the control law of explicit MPC, although the online calculation complexity is reduced, the offline calculation is still prohibitive for complex problems. In this paper, we use the sample points in the feasible region with their corresponding affine functions to construct the lattice PWA approximation of the optimal MPC controller designed for satellite attitude control. The asymptotic stability of satellite attitude control system under lattice PWA approximation has been proven, and simulations are executed to verify that the proposed method can achieve almost the same performance as linear online MPC with much lower online computational complexity and use less fuel than LQR method.
☆ Augmented Labeled Random Finite Sets and Its Application to Group Target Tracking
This paper addresses the problem of group target tracking (GTT), wherein multiple closely spaced targets within a group pose a coordinated motion. To improve the tracking performance, the labeled random finite sets (LRFSs) theory is adopted, and this paper develops a new kind of LRFSs, i.e., augmented LRFSs, which introduces group information into the definition of LRFSs. Specifically, for each element in an LRFS, the kinetic states, track label, and the corresponding group information of its represented target are incorporated. Furthermore, by means of the labeled multi-Bernoulli (LMB) filter with the proposed augmented LRFSs, the group structure is iteratively propagated and updated during the tracking process, which achieves the simultaneously estimation of the kinetic states, track label, and the corresponding group information of multiple group targets, and further improves the GTT tracking performance. Finally, simulation experiments are provided, which well demonstrates the effectiveness of the labeled multi-Bernoulli filter with the proposed augmented LRFSs for GTT tracking.
☆ Integrating Large Language Models for Severity Classification in Traffic Incident Management: A Machine Learning Approach
This study evaluates the impact of large language models on enhancing machine learning processes for managing traffic incidents. It examines the extent to which features generated by modern language models improve or match the accuracy of predictions when classifying the severity of incidents using accident reports. Multiple comparisons performed between combinations of language models and machine learning algorithms, including Gradient Boosted Decision Trees, Random Forests, and Extreme Gradient Boosting. Our research uses both conventional and language model-derived features from texts and incident reports, and their combinations to perform severity classification. Incorporating features from language models with those directly obtained from incident reports has shown to improve, or at least match, the performance of machine learning techniques in assigning severity levels to incidents, particularly when employing Random Forests and Extreme Gradient Boosting methods. This comparison was quantified using the F1-score over uniformly sampled data sets to obtain balanced severity classes. The primary contribution of this research is in the demonstration of how Large Language Models can be integrated into machine learning workflows for incident management, thereby simplifying feature extraction from unstructured text and enhancing or matching the precision of severity predictions using conventional machine learning pipeline. The engineering application of this research is illustrated through the effective use of these language processing models to refine the modelling process for incident severity classification. This work provides significant insights into the application of language processing capabilities in combination with traditional data for improving machine learning pipelines in the context of classifying incident severity.
☆ Adversarial Attacks and Defenses in Automated Control Systems: A Comprehensive Benchmark
Integrating machine learning into Automated Control Systems (ACS) enhances decision-making in industrial process management. One of the limitations to the widespread adoption of these technologies in industry is the vulnerability of neural networks to adversarial attacks. This study explores the threats in deploying deep learning models for fault diagnosis in ACS using the Tennessee Eastman Process dataset. By evaluating three neural networks with different architectures, we subject them to six types of adversarial attacks and explore five different defense methods. Our results highlight the strong vulnerability of models to adversarial samples and the varying effectiveness of defense strategies. We also propose a novel protection approach by combining multiple defense methods and demonstrate it's efficacy. This research contributes several insights into securing machine learning within ACS, ensuring robust fault diagnosis in industrial processes.
☆ Distributed Cooperative Formation Control of Nonlinear Multi-Agent System (UGV) Using Neural Network
The paper presented in this article deals with the issue of distributed cooperative formation of multi-agent systems (MASs). It proposes the use of appropriate neural network control methods to address formation requirements (uncertainties dynamic model). It considers an adaptive leader-follower distributed cooperative formation control based on neural networks (NNs) developed for a class of second-order nonlinear multi-agent systems and neural networks Neural networks are used to compute system data that inputs layer (position, velocity), hidden layers, and output layer. Through collaboration between leader-follower approaches and neural networks with complex systems or complex conditions receive an effective cooperative formation control method. The sufficient conditions for the system stability were derived using Lyapunov stability theory, graph theory, and state space methods. By simulation, the results of this study can be obtained from the main data of the multi-agent system in formation control and verified that the system can process consistency, stability, reliability, and accuracy in cooperative formation.
comment: 6 pages, 5 figures
☆ An Extended Kuramoto Model for Frequency and Phase Synchronization in Delay-Free Networks with Finite Number of Agents
Due to its description of a synchronization between oscillators, the Kuramoto model is an ideal choice for a synchronisation algorithm in networked systems. This requires to achieve not only a frequency synchronization but also a phase synchronization - something the standard Kuramoto model can not provide for a finite number of agents. In this case, a remaining phase difference is necessary to offset differences of the natural frequencies. Setting the Kuramoto model into the context of dynamic consensus and making use of the $n$th order discrete average consensus algorithm, this paper extends the standard Kuramoto model in such a way that frequency and phase synchronization are separated. This in turn leads to an algorithm achieve the required frequency and phase synchronization also for a finite number of agents. Simulations show the viability of this extended Kuramoto model.
comment: 8 pages, 6 figures. Shorter version submitted to the 63rd IEEE Conference on Decision and Control, 2024. Funded by BMBF through 6GEM research hub (16KISK038) and by DFG through project JCRS CoMP (504990291)
☆ Charged Momentum: Electric Vehicle Surge in India's 2023 Landscape
Electric vehicles (EVs) have emerged as a transformative force in India's transportation sector, offering a sustainable solution to the country's growing energy and environmental challenges. Against the backdrop of rapid urbanization, rising pollution levels, and the need for energy security, EVs have gained traction as a viable alternative to traditional internal combustion engine vehicles. This paper provides a comprehensive analysis of the electric vehicle market in India, focusing particularly on the landscape of 2023. It emphasizes key aspects such as the 2023 scenario of EV adoption, the role of indigenous manufacturers, dominant players shaping the market, and the influence of government policies and initiatives, including the FAME I and II schemes. Furthermore, the paper delves into EV sales data for the fiscal year 2023, offering insights into market trends and consumer preferences. By elucidating the current state of EVs in India, this paper aims to contribute to a deeper understanding of the country's transition towards sustainable mobility and its implications for energy, environment, and economy.
☆ Quantifying the Aggregate Flexibility of EV Charging Stations for Dependable Congestion Management Products: A Dutch Case Study
Electric vehicles (EVs) play a crucial role in the transition towards sustainable modes of transportation and thus are critical to the energy transition. As their number grows, managing the aggregate power of EV charging is crucial to maintain grid stability and mitigate congestion. This study analyses more than 500 thousand real charging transactions in the Netherlands to explore the challenge and opportunity for the energy system presented by EV growth and smart charging flexibility. Specifically, it analyses the collective ability to provide congestion management services according to the specifications of those services in the Netherlands. In this study, a data-driven model of charging behaviour is created to explore the implications of delivering dependable congestion management services at various aggregation levels and types of service. The probability of offering specific grid services by different categories of charging stations (CS) is analysed. These probabilities can help EV aggregators, such as charging point operators, make informed decisions about offering congestion mitigation products per relevant regulations and distribution system operators to assess their potential. The ability to offer different flexibility products, namely re-dispatch and capacity limitation, for congestion management, is assessed using various dispatch strategies. Next, machine learning models are used to predict the probability of CSs being able to deliver these products, accounting for uncertainties. Results indicate that residential charging locations have significant potential to provide both products during evening peak hours. While shared EVs offer better certainty regarding arrival and departure times, their small fleet size currently restricts their ability to meet the minimum order size of flexible products.
comment: 18 Pages,13 figures, 2 tables
☆ A Control-Recoverable Added-Noise-based Privacy Scheme for LQ Control in Networked Control Systems
As networked control systems continue to evolve, ensuring the privacy of sensitive data becomes an increasingly pressing concern, especially in situations where the controller is physically separated from the plant. In this paper, we propose a secure control scheme for computing linear quadratic control in a networked control system utilizing two networked controllers, a privacy encoder and a control restorer. Specifically, the encoder generates two state signals blurred with random noise and sends them to the controllers, while the restorer reconstructs the correct control signal. The proposed design effectively preserves the privacy of the control system's state without sacrificing the control performance. We theoretically quantify the privacy-preserving performance in terms of the state estimation error of the controllers and the disclosure probability. Additionally, the proposed privacy-preserving scheme is also proven to satisfy differential privacy. Moreover, we extend the proposed privacy-preserving scheme and evaluation method to cases where collusion between two controllers occurs. Finally, we verify the validity of our proposed scheme through simulations.
☆ A Log-domain Interior Point Method for Convex Quadratic Games
In this paper, we propose an equilibrium-seeking algorithm for finding generalized Nash equilibria of non-cooperative monotone convex quadratic games. Specifically, we recast the Nash equilibrium-seeking problem as variational inequality problem that we solve using a log-domain interior point method and provide a general purpose solver based on this algorithm. This approach is suitable for non-potential, general sum games and does not require extensive structural assumptions. We demonstrate the efficiency and versatility of our method using three benchmark games and demonstrate our algorithm is especially effective on small to medium scale problems.
☆ Observer-Based Environment Robust Control Barrier Functions for Safety-critical Control with Dynamic Obstacles
This paper proposes a safety-critical controller for dynamic and uncertain environments, leveraging a robust environment control barrier function (ECBF) to enhance the robustness against the measurement and prediction uncertainties associated with moving obstacles. The approach reduces conservatism, compared with a worst-case uncertainty approach, by incorporating a state observer for obstacles into the ECBF design. The controller, which guarantees safety, is achieved through solving a quadratic programming problem. The proposed method's effectiveness is demonstrated via a dynamic obstacle-avoidance problem for an autonomous vehicle, including comparisons with established baseline approaches.
☆ Network-Aware Value Stacking of Community Battery via Asynchronous Distributed Optimization
Community battery systems have been widely deployed to provide services to the grid. Unlike a single battery storage system in the community, coordinating multiple community batteries can further unlock their value, enhancing the viability of community battery solutions. However, the centralized control of community batteries relies on the full information of the system, which is less practical and may even lead to privacy leakage. In this paper, we formulate a value-stacking optimization problem for community batteries to interact with local solar, buildings, and the grid, within distribution network constraints. We then propose a distributed algorithm using asynchronous distributed alternate direction method of multipliers (ADMM) to solve the problem. Our algorithm is robust to communication latency between community batteries and the grid while preserving the operational privacy. The simulation results demonstrate the convergence of our proposed asynchronous distributed ADMM algorithm. We also evaluate the electricity cost and the contribution of each value stream in the value-stacking problem for community batteries using real-world data.
comment: 2024 IEEE Power & Energy Society General Meeting (PESGM)
☆ Federated reinforcement learning for robot motion planning with zero-shot generalization
This paper considers the problem of learning a control policy for robot motion planning with zero-shot generalization, i.e., no data collection and policy adaptation is needed when the learned policy is deployed in new environments. We develop a federated reinforcement learning framework that enables collaborative learning of multiple learners and a central server, i.e., the Cloud, without sharing their raw data. In each iteration, each learner uploads its local control policy and the corresponding estimated normalized arrival time to the Cloud, which then computes the global optimum among the learners and broadcasts the optimal policy to the learners. Each learner then selects between its local control policy and that from the Cloud for next iteration. The proposed framework leverages on the derived zero-shot generalization guarantees on arrival time and safety. Theoretical guarantees on almost-sure convergence, almost consensus, Pareto improvement and optimality gap are also provided. Monte Carlo simulation is conducted to evaluate the proposed framework.
☆ Safety-Aware Reinforcement Learning for Electric Vehicle Charging Station Management in Distribution Network
The increasing integration of electric vehicles (EVs) into the grid can pose a significant risk to the distribution system operation in the absence of coordination. In response to the need for effective coordination of EVs within the distribution network, this paper presents a safety-aware reinforcement learning (RL) algorithm designed to manage EV charging stations while ensuring the satisfaction of system constraints. Unlike existing methods, our proposed algorithm does not rely on explicit penalties for constraint violations, eliminating the need for penalty coefficient tuning. Furthermore, managing EV charging stations is further complicated by multiple uncertainties, notably the variability in solar energy generation and energy prices. To address this challenge, we develop an off-policy RL algorithm to efficiently utilize data to learn patterns in such uncertain environments. Our algorithm also incorporates a maximum entropy framework to enhance the RL algorithm's exploratory process, preventing convergence to local optimal solutions. Simulation results demonstrate that our algorithm outperforms traditional RL algorithms in managing EV charging in the distribution network.
comment: 2024 IEEE Power & Energy Society General Meeting (PESGM)
☆ Performance-Guaranteed Solutions for Multi-Agent Optimal Coverage Problems using Submodularity, Curvature, and Greedy Algorithms
We consider a class of multi-agent optimal coverage problems where the goal is to determine the optimal placement of a group of agents in a given mission space such that they maximize a joint ``coverage'' objective. This class of problems is extremely challenging due to the non-convex nature of the mission space and of the coverage objective. With this motivation, we propose to use a greedy algorithm as a means of getting feasible coverage solutions efficiently. Even though such greedy solutions are suboptimal, the submodularity (diminishing returns) property of the coverage objective can be exploited to provide performance bound guarantees - not only for the greedy solutions but also for any subsequently improved solutions. Moreover, we show that improved performance bound guarantees (beyond the standard (1-1/e) performance bound) can be established using various curvature measures that further characterize the considered coverage problem. In particular, we provide a brief review of all existing popular curvature measures found in the submodular maximization literature, including a recent curvature measure that we proposed, and discuss in detail their applicability, practicality, and effectiveness in the context of optimal coverage problems. Moreover, we characterize the dependence of the effectiveness of different curvature measures (in providing improved performance bound guarantees) on the agent sensing capabilities. Finally, we provide several numerical results to support our findings and propose several potential future research directions.
comment: Will be submitted to CDC 2024
☆ A Unified Toll Lane Framework for Autonomous and High-Occupancy Vehicles in Interactive Mixed Autonomy
In this study, we introduce a toll lane framework that optimizes the mixed flow of autonomous and high-occupancy vehicles on freeways, where human-driven and autonomous vehicles of varying commuter occupancy share a segment. Autonomous vehicles, with their ability to maintain shorter headways, boost traffic throughput. Our framework designates a toll lane for autonomous vehicles with high occupancy to use free of charge, while others pay a toll. We explore the lane choice equilibria when all vehicles minimize travel costs, and characterize the equilibria by ranking vehicles by their mobility enhancement potential, a concept we term the mobility degree. Through numerical examples, we demonstrate the framework's utility in addressing design challenges such as setting optimal tolls, determining occupancy thresholds, and designing lane policies, showing how it facilitates the integration of high-occupancy and autonomous vehicles. We also propose an algorithm for assigning rational tolls to decrease total commuter delay and examine the effects of toll non-compliance. Our findings suggest that self-interest-driven behavior mitigates moderate non-compliance impacts, highlighting the framework's resilience. This work presents a pioneering comprehensive analysis of a toll lane framework that emphasizes the coexistence of autonomous and high-occupancy vehicles, offering insights for traffic management improvements and the integration of autonomous vehicles into existing transportation infrastructures.
☆ When are Lossy Energy Storage Optimization Models Convex?
We consider a class of optimization problems involving the optimal operation of a single lossy energy storage system that incurs energy loss when charging or discharging. Such inefficiencies in the energy storage dynamics are known to result in a nonconvex set of feasible charging and discharging power profiles. In this letter, we provide an equivalent reformulation for this class of optimization problems, along with sufficient conditions for the convexity of the proposed reformulation. The conditions provided generalize existing conditions for convexity in the literature.
comment: 5 pages, 1 figure
☆ Sensory Glove-Based Surgical Robot User Interface IROS
Robotic surgery has reached a high level of maturity and has become an integral part of standard surgical care. However, existing surgeon consoles are bulky and take up valuable space in the operating room, present challenges for surgical team coordination, and their proprietary nature makes it difficult to take advantage of recent technological advances, especially in virtual and augmented reality. One potential area for further improvement is the integration of modern sensory gloves into robotic platforms, allowing surgeons to control robotic arms directly with their hand movements intuitively. We propose one such system that combines an HTC Vive tracker, a Manus Meta Prime 3 XR sensory glove, and God Vision wireless smart glasses. The system controls one arm of a da Vinci surgical robot. In addition to moving the arm, the surgeon can use fingers to control the end-effector of the surgical instrument. Hand gestures are used to implement clutching and similar functions. In particular, we introduce clutching of the instrument orientation, a functionality not available in the da Vinci system. The vibrotactile elements of the glove are used to provide feedback to the user when gesture commands are invoked. A preliminary evaluation of the system shows that it has excellent tracking accuracy and allows surgeons to efficiently perform common surgical training tasks with minimal practice with the new interface; this suggests that the interface is highly intuitive. The proposed system is inexpensive, allows rapid prototyping, and opens opportunities for further innovations in the design of surgical robot interfaces.
comment: 6 pages, 5 figures, 7 tables, submitted to International Conference on Intelligent Robots and Systems (IROS)2024
☆ Safety-Aware Perception for Autonomous Collision Avoidance in Dynamic Environments
Autonomous collision avoidance requires accurate environmental perception; however, flight systems often possess limited sensing capabilities with field-of-view (FOV) restrictions. To navigate this challenge, we present a safety-aware approach for online determination of the optimal sensor-pointing direction $\psi_\text{d}$ which utilizes control barrier functions (CBFs). First, we generate a spatial density function $\Phi$ which leverages CBF constraints to map the collision risk of all local coordinates. Then, we convolve $\Phi$ with an attitude-dependent sensor FOV quality function to produce the objective function $\Gamma$ which quantifies the total observed risk for a given pointing direction. Finally, by finding the global optimizer for $\Gamma$, we identify the value of $\psi_\text{d}$ which maximizes the perception of risk within the FOV. We incorporate $\psi_\text{d}$ into a safety-critical flight architecture and conduct a numerical analysis using multiple simulated mission profiles. Our algorithm achieves a success rate of $88-96\%$, constituting a $16-29\%$ improvement compared to the best heuristic methods. We demonstrate the functionality of our approach via a flight demonstration using the Crazyflie 2.1 micro-quadrotor. Without a priori obstacle knowledge, the quadrotor follows a dynamic flight path while simultaneously calculating and tracking $\psi_\text{d}$ to perceive and avoid two static obstacles with an average computation time of 371 $\mu$s.
☆ Credit vs. Discount-Based Congestion Pricing: A Comparison Study
Tolling, or congestion pricing, offers a promising traffic management policy for regulating congestion, but has also attracted criticism for placing outsized financial burdens on low-income users. Credit-based congestion pricing (CBCP) and discount-based congestion pricing (DBCP) policies, which respectively provide travel credits and toll discounts to low-income users on tolled roads, have emerged as promising mechanisms for reducing traffic congestion without worsening societal inequities. However, the optimal design of CBCP and DBCP policies, as well as their relative advantages and disadvantages, remain poorly understood. To address this, we study the effects of implementing CBCP and DBCP policies to route users on a network of multi-lane highways with tolled express lanes. We formulate a non-atomic routing game framework in which a subset of eligible users is granted toll relief in the form of a fixed budget or toll discount, while the remaining ineligible users must pay out-of-pocket. We prove the existence of Nash equilibrium traffic flow patterns corresponding to any given CBCP or DBCP policy. Under the additional assumption that eligible users have time-invariant VoTs, we provide a convex program to efficiently compute these equilibria. For networks consisting of a single edge, we identify conditions under which CBCP policies outperform DBCP policies (and vice versa), in the sense of improving eligible users' access to the express lane. Finally, we present empirical results from a CBCP pilot study of the San Mateo 101 Express Lane Project in California. Our empirical results corroborate our theoretical analysis of the impact of deploying credit-based and discount-based policies, and lend insights into the sensitivity of their impact with respect to the travel demand and users' VoTs.
☆ Sequential Modeling of Complex Marine Navigation: Case Study on a Passenger Vessel (Student Abstract) AAAI 2024
The maritime industry's continuous commitment to sustainability has led to a dedicated exploration of methods to reduce vessel fuel consumption. This paper undertakes this challenge through a machine learning approach, leveraging a real-world dataset spanning two years of a ferry in west coast Canada. Our focus centers on the creation of a time series forecasting model given the dynamic and static states, actions, and disturbances. This model is designed to predict dynamic states based on the actions provided, subsequently serving as an evaluative tool to assess the proficiency of the ferry's operation under the captain's guidance. Additionally, it lays the foundation for future optimization algorithms, providing valuable feedback on decision-making processes. To facilitate future studies, our code is available at \url{https://github.com/pagand/model_optimze_vessel/tree/AAAI}
comment: 5 pages, 3 figures, AAAI 2024 student abstract
☆ Clustering Heuristics for Robust Energy Capacitated Vehicle Routing Problem (ECVRP)
The paper presents an approach to solving the Robust Energy Capacitated Vehicle Routing Problem (RECVRP), focusing on electric vehicles and their limited battery capacity. A finite number of customers, each with their own demand, have to be serviced by an electric vehicle fleet while ensuring that none of the vehicles run out of energy. The time and energy it takes to travel between any two points is modeled as a random variable with known distribution. We propose a Mixed Integer Program (MIP) for computing an exact solution and introduce clustering heuristics to enhance the solution speed. This enables efficient re-planning of routes in dynamic scenarios. The methodology transforms the RECVRP into smaller problems, yielding good quality solutions quickly compared to existing methods. We demonstrate the effectiveness of this approach using a well-known benchmark problem set as well as a set of randomly generated problems.
♻ ☆ Learning Algorithms for Verification of Markov Decision Processes
We present a general framework for applying learning algorithms and heuristical guidance to the verification of Markov decision processes (MDPs). The primary goal of our techniques is to improve performance by avoiding an exhaustive exploration of the state space, instead focussing on particularly relevant areas of the system, guided by heuristics. Our work builds on the previous results of Br{\'{a}}zdil et al., significantly extending it as well as refining several details and fixing errors. The presented framework focuses on probabilistic reachability, which is a core problem in verification, and is instantiated in two distinct scenarios. The first assumes that full knowledge of the MDP is available, in particular precise transition probabilities. It performs a heuristic-driven partial exploration of the model, yielding precise lower and upper bounds on the required probability. The second tackles the case where we may only sample the MDP without knowing the exact transition dynamics. Here, we obtain probabilistic guarantees, again in terms of both the lower and upper bounds, which provides efficient stopping criteria for the approximation. In particular, the latter is an extension of statistical model-checking (SMC) for unbounded properties in MDPs. In contrast to other related approaches, we do not restrict our attention to time-bounded (finite-horizon) or discounted properties, nor assume any particular structural properties of the MDP.
♻ ☆ Secure Synchronization of Heterogeneous Pulse-Coupled Oscillators
In this paper, we consider the synchronization of heterogeneous pulse-coupled oscillators (PCOs), where some of the oscillators might be faulty or malicious. The oscillators interact through identical pulses at discrete instants and evolve continuously with different frequencies otherwise. Despite the presence of misbehaviors, benign oscillators aim to reach synchronization. To achieve this objective, two resilient synchronization protocols are developed in this paper by adapting the real-valued mean-subsequence reduced (MSR) algorithm to pulse-based interactions. The first protocol relies on packet-based communication to transmit absolute frequencies, while the second protocol operates purely with pulses to calculate relative frequencies. In both protocols, each normal oscillator periodically counts the received pulses to detect possible malicious behaviors. By disregarding suspicious pulses from its neighbors, the oscillator updates both its phases and frequencies. The paper establishes sufficient conditions on the initial states and graph structure under which resilient synchronization is achieved in the PCO network. Specifically, the normal oscillators can either detect the presence of malicious nodes or synchronize in both phases and frequencies. Additionally, a comparison between the two algorithms reveals a trade-off between relaxed initial conditions and reduced communication burden.
♻ ☆ Guaranteeing Control Requirements via Reward Shaping in Reinforcement Learning
In addressing control problems such as regulation and tracking through reinforcement learning, it is often required to guarantee that the acquired policy meets essential performance and stability criteria such as a desired settling time and steady-state error prior to deployment. Motivated by this necessity, we present a set of results and a systematic reward shaping procedure that (i) ensures the optimal policy generates trajectories that align with specified control requirements and (ii) allows to assess whether any given policy satisfies them. We validate our approach through comprehensive numerical experiments conducted in two representative environments from OpenAI Gym: the Inverted Pendulum swing-up problem and the Lunar Lander. Utilizing both tabular and deep reinforcement learning methods, our experiments consistently affirm the efficacy of our proposed framework, highlighting its effectiveness in ensuring policy adherence to the prescribed control requirements.
♻ ☆ Unbiased Parameter Estimation via DREM with Annihilators
In the adaptive control theory, the dynamic regressor extension and mixing (DREM) procedure has become widespread as it allows one to describe a variety of adaptive control problems in unified terms of the parameter estimation problem of a regression equation with a scalar regressor. However, when the system/parameterization is affected by perturbations, the estimation laws, which are designed on the basis of such equation, asymptotically provides only biased estimates. In this paper, based on the bias-eliminated least-squares (BELS) approach, a modification of the DREM procedure is proposed that allows one to annihilate perturbations asymptotically and, consequently, asymptotically obtain unbiased estimates. The theoretical results are supported with mathematical modelling and can be used to design adaptive observers and control systems.
comment: 7 pages, 2 figures
♻ ☆ On Continuous Full-Order Integral-Terminal Sliding Mode Control with Unknown Apriori Bound on Uncertainty
This study aims at providing a solution to the problem of designing a continuous and finite-time control for a class of nonlinear systems in the presence of matched uncertainty with an unknown apriori bound. First, we propose a Full-Order Integral-Terminal Sliding Manifold (FOITSM) with a conventional (discontinuous) sliding mode to show that it provides the combined attributes of the nonsingular terminal and integral sliding mode algorithms. Secondly, an Adaptive Disturbance Observer (ADO) has been designed to alleviate the effect of the uncertainty acting on the system. On application of the ADO-based Full-Order Integral-Terminal Sliding Mode Control (FOITSMC), the chattering phenomenon in control input has been reduced substantially in the presence of conditionally known matched disturbances. Moreover, the adaptive gains of ADO are updated non-monotonically without over-bounding the acting disturbance, yet sustain the global boundedness of state trajectories within a specific bound. %Finally, an application of the proposed algorithm for attitude stabilization of a rigid spacecraft has been successively shown.
comment: 14 pages, 9 figures
♻ ☆ Towards Perturbation-Induced Static Pivoting on GPU-Based Linear Solvers
Linear system solving is a key tool for computational power system studies, e.g., optimal power flow, transmission switching, or unit commitment. CPU-based linear system solver speeds, however, have saturated in recent years. Emerging research shows that GPU-based linear system solvers are beginning to achieve notable speedup over CPU-based alternatives in some applications. Due to the architecture of GPU memory access, numerical pivoting represents the new bottleneck which prevents GPU-based solvers from running even faster. Accordingly, this paper proposes a matrix perturbation-based method to induce static pivoting. Using this approach, a series of perturbed, well-conditioned, pivot-free linear systems are solved in parallel on GPUs. Matrix expansion routines are then used to linearly combine the results, and the true solution is recovered to an arbitrarily high degree of theoretical accuracy. We showcase the validity of our approach on distributed-slack AC power flow solve iterations associated with the PGLib 300-bus test case.
comment: submitted to PESGM 2024
♻ ☆ Design of Efficient Point-Mass Filter with Application in Terrain Aided Navigation
This paper deals with state estimation of stochastic models with linear state dynamics, continuous or discrete in time. The emphasis is laid on a numerical solution to the state prediction by the time-update step of the grid-point-based point-mass filter (PMF), which is the most computationally demanding part of the PMF algorithm. A novel efficient PMF (ePMF) estimator, unifying continuous and discrete, approaches is proposed, designed, and discussed. By numerical illustrations, it is shown, that the proposed ePMF can lead to a time complexity reduction that exceeds 99.9% without compromising accuracy. The MATLAB code of the ePMF is released with this paper.
comment: Pulibshed in proccedings of FUSION 2023. PLEASE cite the published version! The code is also now available at GitHub: https://github.com/IDM-UWB/efficient-PMF
RISAR: RIS-assisted Human Activity Recognition with Commercial Wi-Fi Devices
Human activity recognition (HAR) holds significant importance in smart homes, security, and healthcare. Existing systems face limitations because of the insufficient spatial diversity provided by a limited number of antennas. Furthermore, inefficiencies in noise reduction and feature extraction from sensing data pose challenges to recognition performance. This study presents a reconfigurable intelligent surface (RIS)-assisted passive human activity recognition (RISAR) method, compatible with commercial Wi-Fi devices. RISAR leverages a RIS to enhance the spatial diversity of Wi-Fi signals, effectively capturing a wider range of information distributed across the spatial domain. A novel high-dimensional factor model based on random matrix theory is proposed to address noise reduction and feature extraction in the temporal domain. A dual-stream spatial-temporal attention network model is developed to assign variable weights to different characteristics and sequences, mimicking human cognitive processes in prioritizing essential information. Experimental analysis shows that RISAR significantly outperforms existing HAR methods in accuracy and efficiency, achieving an average accuracy of 97.26%. These findings underscore RISAR's adaptability and potential as a robust activity recognition solution in real environments.
♻ ☆ Vehicle Dispatching and Routing of On-Demand Intercity Ride-Pooling Services: A Multi-Agent Hierarchical Reinforcement Learning Approach
The integrated development of city clusters has given rise to an increasing demand for intercity travel. Intercity ride-pooling service exhibits considerable potential in upgrading traditional intercity bus services by implementing demand-responsive enhancements. Nevertheless, its online operations suffer the inherent complexities due to the coupling of vehicle resource allocation among cities and pooled-ride vehicle routing. To tackle these challenges, this study proposes a two-level framework designed to facilitate online fleet management. Specifically, a novel multi-agent feudal reinforcement learning model is proposed at the upper level of the framework to cooperatively assign idle vehicles to different intercity lines, while the lower level updates the routes of vehicles using an adaptive large neighborhood search heuristic. Numerical studies based on the realistic dataset of Xiamen and its surrounding cities in China show that the proposed framework effectively mitigates the supply and demand imbalances, and achieves significant improvement in both the average daily system profit and order fulfillment ratio.
♻ ☆ Contingency Analyses with Warm Starter using Probabilistic Graphical Model
Cyberthreats are an increasingly common risk to the power grid and can thwart secure grid operations. We propose to extend contingency analysis to include cyberthreat evaluations. However, unlike the traditional N-1 or N-2 contingencies, cyberthreats (e.g., MadIoT) require simulating hard-to-solve N-k (with k >> 2) contingencies in a practical amount of time. Purely physics-based power flow solvers, while being accurate, are slow and may not solve N-k contingencies in a timely manner, whereas the emerging data-driven alternatives are fast but not sufficiently generalizable, interpretable, and scalable. To address these challenges, we propose a novel conditional Gaussian Random Field-based data-driven method that performs fast and accurate evaluation of cyberthreats. It achieves speedup of contingency analysis by warm-starting simulations, i.e., improving starting points, for the physical solvers. To improve the physical interpretability and generalizability, the proposed method incorporates domain knowledge by considering the graphical nature of the grid topology. To improve scalability, the method applies physics-informed regularization that reduces model complexity. Experiments validate that simulating MadIoT-induced attacks with our warm starter becomes approximately 5x faster on a realistic 2000-bus system.
comment: arXiv admin note: substantial text overlap with arXiv:2205.03673
♻ ☆ Tropical modeling of battery swapping and charging station
We propose and investigate a queueing model of a battery swapping and charging station (BSCS) for electric vehicles (EVs). A new approach to the analysis of the queueing model is developed, which combines the representation of the model as a stochastic dynamic system with the use of the methods and results of tropical algebra, which deals with the theory and applications of algebraic systems with idempotent operations. We describe the dynamics of the queueing model by a system of recurrence equations that involve random variables (RVs) to represent the interarrival time of incoming EVs. A performance measure for the model is defined as the mean operation cycle time of the station. Furthermore, the system of equations is represented in terms of the tropical algebra in vector form as an implicit linear state dynamic equation. The performance measure takes on the meaning of the mean growth rate of the state vector (the Lyapunov exponent) of the dynamic system. By applying a solution technique of vector equations in tropical algebra, the implicit equation is transformed into an explicit one with a state transition matrix with random entries. The evaluation of the Lyapunov exponent reduces to finding the limit of the expected value of norms of tropical matrix products. This limit is then obtained using results from the tropical spectral theory of deterministic and random matrices. With this approach, we derive a new exact formula for the mean cycle time of the BSCS, which is given in terms of the expected value of the RVs involved. We present the results of the Monte Carlo simulation of the BSCS's operation, which show a good agreement with the exact solution. The application of the obtained solution to evaluate the performance of one BSCS and to find the optimal distribution of battery packs between stations in a network of BSCSs is discussed.
comment: 21 pages, 5 figures
♻ ☆ Application of tropical optimization for solving multicriteria problems of pairwise comparisons using log-Chebyshev approximation
We consider a decision-making problem to find absolute ratings of alternatives that are compared in pairs under multiple criteria, subject to constraints in the form of two-sided bounds on ratios between the ratings. Given matrices of pairwise comparisons made according to the criteria, the problem is formulated as the log-Chebyshev approximation of these matrices by a common consistent matrix (a symmetrically reciprocal matrix of unit rank) to minimize the approximation errors for all matrices simultaneously. We rearrange the approximation problem as a constrained multiobjective optimization problem of finding a vector that determines the approximating consistent matrix. The problem is then represented in the framework of tropical algebra, which deals with the theory and applications of idempotent semirings and provides a formal basis for fuzzy and interval arithmetic. We apply methods and results of tropical optimization to develop a new approach for handling the multiobjective optimization problem according to various principles of optimality. New complete solutions in the sense of the max-ordering, lexicographic ordering and lexicographic max-ordering optimality are obtained, which are given in a compact vector form ready for formal analysis and efficient computation. We present numerical examples of solving multicriteria problems of rating four alternatives from pairwise comparisons to illustrate the technique and compare it with others.
comment: 44 pages